Apps Script · Web app
Invoicing and driver pay, built on Google Sheets
6 versions shipped · 250+ loads tracked
Live aging dashboard, a 16-line invoice form, PDF invoices to Drive and per-driver pay reports.
Watch this first · 40 secondsHi · A 40-sec intro for the team
Stop doing it by hand.
Reports, invoices, payroll, data entry, documents. I turn the work your team repeats every week into automations that run on their own, in the tools you already use.
40-sec introSound on 🔊
Get my free automation plan
2-minute formFree 20-min callFixed quote, no obligation
5.0 from 144 client reviews
178 projects shipped
7 years in finance (FP&A, Big-4)
upworkTop Ratedfiverr.Top Rated
Built inside the tools you already use
- Sheets
- Apps Script
- Excel
- Gmail
- Drive
- Looker Studio
- BigQuery
- Power BI
- Forms
- Calendar
- Docs
- Outlook
- Word
- Python
- OpenAI
- Claude
- Gemini
- Airtable
- Supabase
- React
- Next.js
- Make
- n8n
- HubSpot
- monday
- ClickUp
Sound familiar?
Your team is too good to spend its week on copy‑paste.
Copy-pasting between tabs every MondayExports, VLOOKUPs, the same twelve clicks, week after week.
The report that takes a day to buildAnd it's already out of date by the time anyone reads it.
Invoices and pay worked out by handTyped, checked, re-checked, then chased by email.
Emails and PDFs retyped into a sheetOrders, receipts, requests: someone reads each one and types it in.
Approvals and follow-ups chased by handReminders, status updates and sign-offs living in someone's inbox.
One person who “knows how the file works”And everything stops when they're on holiday.
Every one of these is a solved problem. Here's the proof, from real clients.
Real work, real clients
Fifteen jobs that used to be done by hand.
Each one started with a message like “we do this every week, can it be automated?”. Client names are hidden. The numbers are real.
AI · Document pipeline
Scanned receipts, split and renamed by AI
32 scans in · 34 receipts out · 0 failed
Watches every client folder in SharePoint, splits pages with several receipts and names each file date_supplier_amount.
Web app · Payroll
Timesheets and payslips for a moving company
Bilingual · row-level security
Daily hours in, allowances calculated, payslip PDFs out. Built with Next.js and Supabase.
AI · Voice to data
Phone pledges turned into spreadsheet rows
No manual typing left
Calls transcribed, then a language model extracts donor, amount and campaign into Google Sheets.
Forms · Live dashboard
One live view of every coach's lessons
Monthly pay straight from the data
A 30-second form for coaches, Apps Script clean-up, and a dashboard that shows who hasn't reported.
Mobile app · Dashboard
Shelf audits moved from Google Forms to a mobile app
31 stores · 1,399 photos migrated
Four-step phone audits, role-based security and a live stock-out dashboard on Supabase.
Apps Script · Rescue
Tutor requests from Gmail, filed into the right row
Broken intake rebuilt on V8
A label-driven parser, clean triggers and a Drive file per tutor, documented for a non-technical owner.
Apps Script · Calendar
Every free slot in the calendar, found and sorted
“Super helpful and works great!”
Reads any date range, finds the gaps in working hours and groups them by length in Sheets.
Browser automation · Portal
New court filings, waiting every Monday
Weekly · private two-login portal
Browser automation county by county, plus a portal with a countdown and one-click CSV.
Long-running pipeline
UK property listings, delivered weekly for 4 years
2021 → 2025 · every week
Listings standardised and compared with the last run, fixed fast whenever a portal changed.
Data verification
10,000+ UK companies checked against the register
10,000+ records verified
Batch checks against the official register, dissolved companies flagged, gaps filled from public sources.
Registry · Extraction
The owners of Swiss companies, from official registries
Resumable, cached runs
Commercial-register API plus gazette parsing to list every officer, role and signing right.
Data pipeline · Client portal
Every new UCC filing, collected and matched daily
136,118 filings matched
Collectors on official state data, fuzzy matching against 156 lenders, and a branded portal with filters and export.
Data sync · CRM
New property listings in the CRM every morning
Daily · zero manual steps
New owner listings cleaned, de-duplicated against the CRM and imported for the call team before 9am.
Data quality · Enrichment
328 coffee shops, enriched and verified
16 cities · 70 bad records caught
Collected from Google Places, enriched with site, phone, email and socials, every row checked.
Your process is probably simpler than most of these.
Tell me about it in 2 minutes. I'll tell you if it's worth automating.
How it works
From “we do this by hand” to “it just runs”, in three steps.
Tell me what you do by hand
Fill in the 2-minute form, then we talk for 20 minutes. Share your screen, show me the file. No preparation needed.
Free · 20 minutesGet a plan and one fixed price
I map the process, the edge cases and the tools, then send a written plan with a fixed quote. You decide, with no pressure.
Within 48 hoursIt runs. You get your hours back.
I build, test on your real data, hand over with a short video and written notes, and stay reachable when things change.
Usually 1–3 weeksYou own everything: code, files and accounts
Built in tools you already pay for
Video + written handover
A direct line to the person who built it
Start with the 2-minute form
FreeNo obligation
Hi, I'm LateefWho you'll work with
A finance guy who learned to code. Not the other way round.
I spent seven years in FP&A and Big-4 finance, building the reports everyone else copied and pasted. Now I build what replaces the copy-paste: Google Sheets and Excel apps, AI document pipelines, small web apps and live dashboards.
Because I come from finance, I ask about the numbers first: how many hours, how many errors, what it costs you. If an automation won't pay for itself, I'll tell you.
178 projects shipped144 client reviewsEnglish · Français · العربيةWorldwide · any time zone
upworkTop Ratedfiverr.Top Rated
What clients say
Short version: it works.
“Helped us build a custom script in Google Calendar ↔ Google Sheets… Super helpful and works great!”Services client · United States
“Succinct and professional, and did my job very quickly and thoroughly. I would use him again.”Small-business client
178projects shipped
144client reviews
< 24 hreply time, in any time zone
Ongoingsupport after handover
NDAsigned on request, your data stays private
Still looking for your case?
120 builds. Find the one that looks like your problem.
Search by tool or task: invoices, Power Query, Gmail, AI, payroll…
The library
120 builds, 47 for clients.
No build matches that yet.
That doesn't mean I can't build it. Tell me what you need.
Tradeforce
Rescuing a 165,000-row sheet from recalculation collapse
165k+rows, recalc near-instant
Array-formula recalculation had made the workbook unusable. The compute moved out of Sheets into an offline Python ETL that rebuilds the dataset and writes it back in batched, quota-aware API calls.
Under the hood
- Volatile array formulas replaced by precomputed columns
- Chunked batchUpdate writes with exponential backoff
- Idempotent re-runs: same input, same output
- Python
- Sheets API
- Pandas
- ETL
Finance team · Switzerland
Month-end close compressed from 3 days to 3 hours
3d → 3hmonth-end reporting
Twelve source exports, three ledgers and one reporting pack: Power Query ingests, cleans and reconciles everything, and the full pack rebuilds itself on refresh.
Under the hood
- Parameterised M queries with schema-drift handling
- Data model with mapped chart of accounts
- Built-in reconciliation checks before publishing
- Excel
- Power Query
- Reporting
- Data modeling
Government operations unit · UAE
A 10-sheet ops system under a strict no-VBA, no-BI constraint
10linked sheets, 1 dashboard
Government rules banned macros and Power BI, so the whole operational logic lives in pure formulas: dynamic CSE arrays across ten linked sheets feeding a live dashboard.
Under the hood
- Cross-sheet CSE array engine, zero macros
- Validation layers that block bad entries at source
- Dashboard recalculates in real time on 2,000+ records
- Excel
- Array formulas
- Dashboard
- Data validation
Nonprofit donation line · US
Voice-to-database pipeline with LLM extraction
< 8scall to structured row
A serverless pipeline answers the call, transcribes it with Whisper, extracts donor, amount and intent with a structured-output LLM prompt, and writes a validated row to Sheets in under 8 seconds.
Under the hood
- Twilio webhooks → serverless function
- JSON-schema LLM extraction with confidence checks
- Failed parses routed to human review
- Twilio
- Whisper
- GPT
- Google Sheets
- Serverless
- Webhooks
BLU Cleaning
Rules-based payroll engine for a 48-person workforce
0manual payroll math
Shift hours, overtime tiers, rates and deductions are calculated by an Apps Script engine every cycle, with an audit trail for every number on every payslip.
Under the hood
- Configurable pay rules, no hard-coded rates
- Overtime and edge cases unit-tested
- Immutable payroll snapshots per cycle
- Google Sheets
- Apps Script
- Audit trail
Enterprise finance (FP&A)
Enterprise reporting automation for Fortune 500 finance teams
F500teams using it
Period-end management reporting that analysts used to rebuild by hand, re-engineered as an automated pipeline from ERP extracts to a governed Power BI model.
Under the hood
- Star-schema data model with DAX measures
- Automated variance and commentary tables
- Refresh-ready for every reporting period
- Excel
- Power BI
- DAX
- ERP data
Services business · US
Calendar intelligence: free-time detection and bucketing
★ 5.0“Super helpful and works great!”
A Calendar ↔ Sheets engine reads every event across calendars, computes the gaps between them, handles time zones and all-day events, and buckets the free time by length.
Under the hood
- Interval-merging algorithm across calendars
- Time-zone and recurring-event handling
- Time-driven triggers keep it current
- Apps Script
- Google Calendar
- Google Sheets
- Algorithms
Multi-store retail · EU
Multi-store KPI warehouse with a live executive dashboard
1 viewevery store, live
Point-of-sale data from every store is loaded into BigQuery, modelled into clean KPI tables and served through one Looker Studio dashboard that refreshes on its own.
Under the hood
- Scheduled loads with partitioned tables
- SQL transformation layer for KPIs
- Store-level drill-down from one view
- BigQuery
- Looker Studio
- SQL
- Data warehouse
E-commerce brand · UK
Two-way inventory sync across Shopify, suppliers and ERP sheet
0stockouts from sync gaps
Shopify, several supplier feeds and a master sheet stay in continuous two-way sync, with conflict resolution so the same item is never wrong in two places.
Under the hood
- Delta sync with last-write-wins + overrides
- Supplier feed parsers (CSV, XML, API)
- Sync health monitoring and alerts
- Shopify API
- Apps Script
- Google Sheets
- Conflict resolution
Willow Glen Tutoring
Zero-downtime migration of a 21-file legacy codebase (Rhino → V8)
21 → 1script files
A tutoring business's whole Apps Script automation, spread over 21 files, was migrated off Google's deprecated Rhino runtime and refactored into one modular V8 codebase while it stayed live. The broken Request-a-Tutor intake was rebuilt too: a Gmail-label parser that writes each request to the right row instead of overwriting students.
Under the hood
- Runtime incompatibilities found and patched, orphaned triggers removed
- Label-driven Gmail parser with row-matching, no overwrites
- Cut-over with no lost triggers or data; owner-deployable docs
- Apps Script
- V8
- Gmail
- JavaScript
- Refactoring
- Gmail API
Hicham Ritab — Real Estate
Institutional-grade real-estate underwriting model
1 modelevery deal, same framework
One reusable model runs every deal through the same assumptions: financing, cash flows, sensitivity tables and scenario outputs. Deals are comparable side by side.
Under the hood
- Scenario manager with base/bull/bear cases
- IRR, cash-on-cash and DSCR outputs
- Locked logic, open inputs
- Google Sheets
- Financial modeling
- Underwriting
Services SMB · EU
Invoice OCR pipeline with field-level validation
secsinvoice → ledger row
Incoming invoices are OCR'd, key fields are extracted and checked (totals vs. line items, VAT, dates), and a structured row is posted to the ledger within seconds.
Under the hood
- OCR + layout parsing per vendor template
- Arithmetic validation before posting
- Exceptions queue for low-confidence reads
- Python
- OCR
- Google Sheets
- Document AI
Toyota dealership
Competitive price-intelligence system for a car dealership
15 mindetection loop
Scheduled crawlers watch competitor inventory and pricing; a change-detection layer filters out noise and alerts sales the moment a relevant price moves.
Under the hood
- Headless crawling with anti-bot resilience
- Change detection with noise thresholds
- 15-minute monitoring loop
- Python
- Scheduler
- Email alerts
- Change detection
Fitness industry client · US
Geospatial competitor intelligence: 1,398 venues mapped
1,398gyms mapped & QA'd
A national dataset was built, QA'd record by record and geocoded, then shipped as an interactive Leaflet map with clustering and filters.
Under the hood
- Geocoding with confidence scoring
- Multi-pass data QA and correction
- Clustered interactive map layer
- Python
- Data QA
- Leaflet
- Geocoding
Hospitality · France
Event-driven booking automation with reminder orchestration
0no-shows chased by hand
Every booking triggers a chain: calendar event, confirmation, then timed reminders, with reschedules and cancellations handled cleanly across the whole sequence.
Under the hood
- Event-driven triggers, no polling
- State machine for reschedule/cancel
- Templated multi-channel messaging
- Google Calendar
- Apps Script
- Gmail
- Orchestration
Freight & logistics client
End-to-end carrier performance analytics
Shipment data from multiple carriers is consolidated, normalised and scored on on-time rate, delays and cost, then published as an automated weekly scorecard.
Under the hood
- Carrier data normalisation layer
- KPI engine with weighted scoring
- Scheduled report generation
- Apps Script
- Google Sheets
- Reporting
- KPI design
Retail client
Real-time stock control system with reorder logic
Every stock movement is logged; live levels, reorder points and lead times drive automatic reorder flags. The warehouse runs on it daily.
Under the hood
- Movement ledger, not overwritten totals
- Dynamic reorder-point calculation
- Role-protected input ranges
- Google Sheets
- Formulas
- Inventory logic
Operations client
40-step manual process collapsed into one VBA command
A complex VBA module consolidates dozens of sheets, cleans and formats the data, and generates the finished report in a single click, with error handling at every stage.
Under the hood
- Modular VBA with structured error handling
- Array-based processing for speed
- Logging of every run
- Excel
- VBA
- Macros
- Error handling
Property data client · UK
Long-running property data pipeline (4+ years in production)
4+ yrsrunning for one client
A production scraper has collected UK property-portal listings for over four years, surviving site redesigns and delivering clean, diffed datasets on schedule.
Under the hood
- Resilient selectors and layout-change alerts
- Run-over-run diffing (new / changed / removed)
- Four years of uptime for one client
- Python
- Scraping
- Scheduler
- Production ops
Business-intelligence client
Financial-products dataset acquired from Bankrate & WalletHub
★ 5.05.0 ★ client review
Rates and product terms are extracted at scale from two major comparison sites and normalised into one comparable dataset for business-intelligence work.
Under the hood
- Dynamic-page extraction
- Cross-source schema harmonisation
- Validation against expected ranges
- Python
- Scraping
- Data cleaning
- Data harmonisation
Operations client
Multi-file consolidation into a single source of truth
Dozens of independently maintained sheets with different layouts are mapped, normalised and compiled into one master that stays current automatically.
Under the hood
- Column-mapping layer for mismatched schemas
- Incremental import, not full rebuilds
- Source tracking on every row
- Google Sheets
- Apps Script
- Data integration
Small business client
Advanced analytical workbook with macros and dynamic charts
★ 5.0“Very quickly and thoroughly”
A spreadsheet engine combining advanced formulas, VBA macros and dynamic charts: delivered fast and thoroughly.
Under the hood
- Dynamic named ranges
- Macro-driven refresh and formatting
- Charts bound to live calculations
- Excel
- VBA
- Charts
- Dynamic arrays
Accounting client
Automated bank reconciliation with fuzzy matching
Bank transactions are matched to receipts by amount, date window and fuzzy vendor names. Split payments and duplicates are handled; only true exceptions reach a human.
Under the hood
- Multi-key fuzzy matching algorithm
- Tolerance windows for dates and amounts
- Exception report for unmatched items
- Python
- Excel
- Matching logic
- Reconciliation
Accounting firm · Switzerland
Vision-AI receipt splitting and renaming pipeline on SharePoint
0 failed32 scans → 34 named receipts
A Swiss accounting firm's scans are picked up from each client's SharePoint folder; every receipt is detected (several per page, up to 64-page scans), split into its own PDF and named date_supplier_currency_amount by a vision model. Low-confidence results go to review and originals are archived, all on the firm's own server.
Under the hood
- Multi-receipt detection and page segmentation
- Vision-model field extraction with confidence-based review queue
- Web dashboard with login and per-client switches
- Python
- Vision AI
- SharePoint
- Computer vision
- Self-hosted
Professional services client
High-volume document generation engine (Excel → PDF)
Every row of a data file becomes a fully formatted, personalised PDF. Hundreds are generated in one run with consistent layout and naming.
Under the hood
- Template engine with conditional sections
- Batch rendering with error isolation
- Output naming and folder routing
- Excel
- Python
- Templating
Trucking company · US
Driver compensation and invoicing platform on Apps Script
250+loads invoiced & paid
A web app built on the company's Google Sheet replaces a slow, messy invoicing file: a live dashboard (invoiced, collected, outstanding, aging), a 16-line invoice form with live totals, PDF invoices filed to Drive, driver pay and per-driver reports. Six versions shipped as the business grew.
Under the hood
- Pay rules engine per load type, with per-driver reports
- Audit trail on every invoice edit and regeneration
- Privacy switch that hides amounts for screen sharing
- Apps Script
- Google Sheets
- Web app
- Business logic
- Audit trail
Moving company · Sweden
Bilingual timesheet and payroll platform with row-level security
2languages, one payroll engine
A Swedish moving company ran hours, allowances and pay through monthly spreadsheets. It now runs on a Next.js + Supabase app: daily hours per employee, allowance and overtime rules applied automatically, and a payslip PDF generated per person, in Swedish or English.
Under the hood
- Postgres row-level security: each user sees only their own scope
- Allowance and overtime rules computed server-side, not in the browser
- i18n layer switching every label, payslip and export between SV and EN
- Next.js
- Supabase
- Postgres
- i18n
Logistics startup
Transport management system recovered and shipped to production
A half-built TMS with a broken build and tangled state was diagnosed, refactored and brought to a stable, deployable product.
Under the hood
- Root-cause analysis of build failures
- State and data-flow refactor
- Production deployment pipeline
- React
- Supabase
- TMS
- CI/CD
Sports academy
Coach performance and lesson-compliance dashboard
Lesson submissions from every coach flow into a live dashboard that shows compliance, gaps and trends for the academy owner.
Under the hood
- Form → sheet ingestion with validation
- Compliance logic per coach and period
- Live dashboard with filters
- Google Forms
- Google Sheets
- Looker Studio
- Dashboard
E-commerce client
Cross-platform workbook engineered for Excel and Google Sheets
One workbook built to behave identically in both platforms, using only functions with matching semantics and avoiding each platform's traps.
Under the hood
- Function-compatibility matrix
- Platform-safe formula patterns
- Tested in both environments
- Excel
- Google Sheets
- Formulas
- Compatibility
SMB client
Technical audit of a business-critical Google Sheets system
A structured audit of a mission-critical Sheets setup: broken references, performance bottlenecks and data-loss risks found, ranked by impact and scoped for fixing.
Under the hood
- Dependency and reference mapping
- Performance profiling of heavy formulas
- Risk-ranked remediation plan
- Google Sheets
- Audit
- Apps Script
Operations client
API integration layer between Google Sheets and business systems
Apps Script services connect Sheets to external platforms over REST: authentication, pagination, rate limits and scheduled syncs handled properly.
Under the hood
- OAuth / API-key auth handling
- Pagination and rate-limit backoff
- Scheduled bidirectional sync
- Apps Script
- REST APIs
- Google Sheets
- OAuth
Data client · UK
Company data verification application
An app checks company records against official registry sources: status, registration numbers and addresses. It flags every mismatch before it causes damage.
Under the hood
- Registry API cross-checks
- Rule-based mismatch detection
- Bulk verification with reports
- Python
- APIs
- Web app
- Data quality
SaaS client
Application data migration to Supabase/Postgres
An application was moved off fragile storage onto a relational Supabase/Postgres schema, with data migrated, integrity enforced and the app re-wired to the new backend.
Under the hood
- Relational schema design with constraints
- Scripted, repeatable migration
- Row-level security policies
- Supabase
- Postgres
- SQL
- RLS
Confidential documents client
Per-recipient document encryption and forensic watermarking
Every outgoing document is watermarked with the recipient's identity and encrypted, so any leak can be traced back to its source.
Under the hood
- Per-recipient visible + invisible marks
- PDF encryption with permissions
- Batch distribution workflow
- Python
- Encryption
- Security
Retail field team
Mobile shelf-audit app feeding a central data model
31stores audited from phones
A skincare brand's Google Form was replaced by a phone-first app across 31 stores: reps photograph the shelf and flag missing or low products in four steps, and managers see availability and repeat stock-outs live. 1,247 past responses and 1,399 photos were migrated in.
Under the hood
- Supabase with role-based security and photo storage
- Four-step mobile capture flow per store visit
- Historical migration from Google Forms and Drive
- Mobile app
- Google Sheets
- Forms
- Mobile
- Supabase
Demo build · price intelligence
Cross-provider remittance price intelligence
Live quotes from money-transfer providers are collected per corridor and amount, normalised for fees and FX margin, and ranked to show the real cheapest option.
Under the hood
- Quote collection across providers
- Fee + FX-margin normalisation
- Corridor-level ranking
- Python
- Scraping
- Dashboard
- FX
Demo build · deal finder
Arbitrage engine for trading-card marketplaces
New marketplace listings are scanned against a market-price model; underpriced cards trigger an instant alert before other buyers see them.
Under the hood
- Market-price model from sold listings
- Under-market scoring
- Real-time push alerts
- Python
- Scraping
- Alerts
- Pricing model
Demo build · data engineering
Sports analytics warehouse: ten seasons in SQL
Match, player and league-table data is modelled into a normalised SQLite schema, ready for any analysis, model or dashboard.
Under the hood
- Normalised relational schema
- ETL from multiple sources
- Analytical SQL views
- Python
- SQLite
- SQL
- Data modeling
Demo build · API integration
Wearable-to-calendar sync over OAuth APIs
Sleep, recovery and workout data is pulled from a wearable's OAuth API and written into Google Calendar as structured events every night.
Under the hood
- OAuth 2.0 token refresh flow
- Data → event mapping
- Nightly scheduled sync
- API
- OAuth
- Google Calendar
Demo build · alerts
AI-scored job-opportunity radar
New job posts are pulled from an API, filtered by rules and scored by an LLM for fit, and only high-score opportunities are pushed to you in real time.
Under the hood
- API polling with de-duplication
- LLM fit-scoring
- Real-time notifications
- Python
- API
- Alerts
- LLM
Demo build · calculator
Interactive marketing ROI simulator
A scenario model projects leads, revenue and ROI from campaign inputs, used live on sales calls to show prospects the numbers.
Under the hood
- Driver-based projection model
- Scenario toggles
- Client-facing presentation layer
- Google Sheets
- Formulas
- Modeling
Demo build · HR
Compensation tracker with history and export engine
Salaries, raises and full history in one governed sheet, with one-click exports to formatted PDF and Excel.
Under the hood
- Effective-dated salary history
- Protected sensitive ranges
- Export engine (PDF / XLSX)
- Google Sheets
- Apps Script
- Governance
Demo build · finance
Self-updating 13-week cash-flow forecast
Bank exports land in a folder; the forecast refreshes, rolls forward and highlights the weeks where cash gets tight.
Under the hood
- Rolling 13-week direct-method model
- Power Query bank ingestion
- Low-cash early warnings
- Excel
- Power Query
- Forecasting
- Treasury
Demo build · FP&A
Board-ready budget-vs-actual variance engine
Actuals and budget are mapped through the chart of accounts and every variance is calculated and explained line by line, ready for the board.
Under the hood
- Account mapping layer
- Price / volume / mix variance
- Automated commentary tables
- Excel
- Power Pivot
- Reporting
- FP&A
Demo build · month-end close
Automated balance-sheet reconciliation workbook for month-end close
46accounts tied out each close
GL trial balance and sub-ledger exports (AR, AP, fixed assets, payroll, bank) are pulled in by Power Query and every one of 46 balance-sheet accounts is tied out on refresh. Differences above a materiality threshold must carry an explanation before the reviewer can sign off.
Under the hood
- Account-to-sub-ledger mapping table drives every tie-out
- Materiality thresholds per account with mandatory commentary
- Preparer / reviewer sign-off with timestamped audit trail
- Excel
- Power Query
- Reconciliation
- Month-end close
- Audit trail
Demo build · month-end close
Dependency-gated month-end close orchestrator
WD+4close, down from WD+9
A 140-task close checklist modelled as a dependency graph: each task has an owner, a workday deadline relative to period end and its predecessors. Apps Script unlocks tasks as predecessors complete, chases late owners and publishes a live close-status board.
Under the hood
- Task dependency graph with critical-path highlighting
- Workday arithmetic that skips weekends and local holidays
- Escalating reminders and a live status board for the controller
- Google Sheets
- Apps Script
- Gmail
- Workflow
- Critical path
Demo build · month-end close
GRNI accrual engine driven by PO and goods-receipt data
1,800PO lines assessed monthly
Instead of emailing budget holders for estimates, the engine matches 1,800 open PO lines against goods receipts and posted invoices, accrues only what was received and not yet billed, and auto-reverses last month's accrual.
Under the hood
- Three-way join of POs, receipts and invoices at line level
- Partial-receipt and price-variance handling
- Auto-reversing journal output in ERP upload format
- Excel
- Power Query
- ERP exports
- Accruals
- ERP
Demo build · management accounting
Driver-based shared-cost allocation engine
14cost centres, one refresh
Central costs are allocated through configurable driver tables (headcount, m², revenue, ticket volume), with step-down allocation between service departments and a fully balanced journal that proves nothing was lost or double-counted.
Under the hood
- Configurable driver matrix per cost pool
- Step-down allocation between service departments
- Zero-sum proof check before the journal is released
- Excel
- Power Pivot
- DAX
- Cost accounting
Demo build · accounting
Fixed-asset register with multi-method depreciation and roll-forward
2,310assets depreciated monthly
A governed register for 2,310 assets runs straight-line, declining-balance and mid-month conventions per asset class, handles disposals with gain/loss calculation, and produces a cost and accumulated-depreciation roll-forward that ties to the GL.
Under the hood
- Per-class depreciation methods and useful lives
- Disposal gain/loss and partial-period logic
- Roll-forward reconciled to GL every month
- Excel
- Power Query
- Formulas
- Fixed assets
Demo build · SaaS finance
ASC 606 / IFRS 15 revenue-recognition and deferred-revenue engine
1,150contracts on schedule
Billing exports for 1,150 subscription contracts are turned into a monthly recognition schedule: daily proration, mid-term upgrades, cancellations and multi-element bundles allocated by standalone selling price, with a deferred-revenue waterfall that ties to the balance sheet.
Under the hood
- SSP allocation across bundled performance obligations
- Daily proration with mid-term contract modifications
- Waterfall reconciled to deferred-revenue GL balance
- Excel
- Power Query
- Revenue recognition
- Rev rec
- SaaS
Demo build · lease accounting
AI lease abstraction feeding an IFRS 16 accounting engine
38leases abstracted & scheduled
An LLM extracts commencement date, term, payment schedule, escalations and options from each lease PDF into a validated abstract; the engine then discounts the payments at the incremental borrowing rate and generates ROU, liability and interest schedules with monthly journals.
Under the hood
- Schema-validated lease abstracts with clause references
- Present-value engine with IBR and indexation remeasurement
- Journal drafts and disclosure tables generated per period
- Python
- LLM
- Excel
- IFRS 16
- Document AI
Demo build · AI for finance
LLM-assisted month-end flux analysis
Month-over-month and year-over-year movements are computed per account; anything above a dual absolute/percentage threshold has its GL lines ranked by contribution, and an LLM drafts the variance explanation from that evidence only, with citations to the underlying entries.
Under the hood
- Dual absolute + % thresholds per account class
- Top-contributor decomposition from journal-line detail
- Grounded drafts that cite source entries; reviewer approves
- Python
- LLM
- Excel
- Variance analysis
Demo build · Office Scripts
Pre-ERP journal validation pipeline on Office Scripts + Power Automate
0rejected ERP uploads
Journals are prepared in an Excel Online template; an Office Script validates balance, account/cost-centre combinations, open periods and approval limits against master data, and Power Automate routes the clean file to the right approver before submission.
Under the hood
- TypeScript Office Scripts validating against master-data tables
- Approval routing by amount and entity in Power Automate
- Blocked lines returned to the preparer with reasons
- Excel Online
- Office Scripts
- Power Automate
- TypeScript
- Approvals
Demo build · group reporting
Group chart-of-accounts mapping layer across three ERPs
1,240 → 186accounts mapped
1,240 local accounts from SAP, Sage and a legacy ledger are mapped to a 186-line group chart through a governed mapping table; unmapped or newly created accounts are caught on every refresh before they can distort group reporting.
Under the hood
- Versioned mapping table with effective dates
- Unmapped-account detection on every refresh
- Entity-level tie-out of local TB to group TB
- Excel
- Power Query
- Data modeling
- Master data
Demo build · FP&A
Driver-based three-statement financial model
3statements, fully linked
A five-year monthly model where volume, price, headcount and working-capital drivers flow into a fully linked P&L, balance sheet and indirect cash flow. A circularity-safe interest calculation and balance checks on every period keep it bankable.
Under the hood
- Working-capital drivers (DSO, DPO, DIO) feeding cash flow
- Circularity-safe interest with a copy-paste breaker
- Scenario switch plus balance and cash checks every period
- Excel
- Financial modeling
- Scenarios
Demo build · FP&A
Rolling 12-month reforecast with automated actuals roll-in
12 moalways-on forecast horizon
Each month the closed period's actuals replace forecast through Power Query, the horizon rolls forward, and run-rate, seasonality and pipeline-based methods are blended per line to produce the updated full-year outlook and forecast-accuracy tracking.
Under the hood
- Actuals-over-forecast replacement by period flag
- Method selection per line: run-rate, seasonal, pipeline
- Forecast-accuracy (MAPE) tracked version over version
- Excel
- Power Query
- Forecasting
Demo build · workforce planning
Position-level workforce planning and fully loaded cost model
340positions planned
Each current and planned position carries its start date, band, FTE and cost centre; the model applies salary increases, employer taxes, benefits and bonus accruals month by month and reconciles plan headcount to the HRIS every refresh.
Under the hood
- Position-level (not average) cost build
- Employer tax and benefit rules by country
- Plan-to-HRIS reconciliation of filled vs. open roles
- Excel
- Power Query
- HRIS export
- Workforce planning
Demo build · budgeting
Distributed budget-collection workflow on Google Workspace
18departments, zero emailed files
Apps Script generates a pre-filled budget file per department with prior-year actuals and guardrails, shares it with the owner, validates submissions, locks files on deadline and consolidates every version into a master budget with change tracking.
Under the hood
- Template generation and permissions via Drive API
- Submission validation and deadline locks
- Versioned consolidation with change log per department
- Google Sheets
- Apps Script
- Drive API
- Budgeting
Demo build · treasury
Debt schedule and covenant-compliance forecasting model
Every facility's drawdowns, amortisation, interest (fixed and floating with rate curves) and fees are modelled monthly; covenant ratios are computed on the bank's own definitions and projected forward against the forecast so breaches surface quarters early.
Under the hood
- Facility-level amortisation with floating-rate curves
- Covenant definitions modelled exactly as in the agreement
- Forward headroom testing against the rolling forecast
- Excel
- Financial modeling
- Forecasting
- Treasury
- Debt
Demo build · treasury
Daily multi-bank cash-positioning pipeline
9banks, zero portal logins
Bank statement files (MT940, CAMT.053 and CSV) from nine banks are collected and parsed each morning, converted to the group currency, categorised and loaded into a cash-position view by entity, bank and currency before the treasury call.
Under the hood
- MT940 / CAMT.053 statement parsers
- FX conversion and intraday-vs-closing balance logic
- Missing-feed detection before the 8:00 cut-off
- Python
- Bank APIs
- Google Sheets
- Treasury
- ETL
Demo build · sales
CPQ-lite: quote generation from a product sheet
Products and pricing rules live in a sheet; one click configures the quote, applies discounts and produces a branded PDF that is emailed out.
Under the hood
- Pricing and discount rules
- Docs template rendering
- Quote numbering and log
- Google Sheets
- Google Docs
- CPQ
Demo build · workflow
Multi-level purchase approval workflow
Requests are routed by amount and department to the right approver, approved from Gmail with one click and logged end to end.
Under the hood
- Approval routing matrix
- Actionable email approvals
- Full audit log
- Google Sheets
- Gmail
- Apps Script
- Workflow
Demo build · cash collection
Accounts-receivable automation with dunning sequences
Invoices go out on schedule; escalating reminder sequences follow automatically until each one is paid.
Under the hood
- Recurring invoice scheduler
- Escalating dunning cadence
- Payment status tracking
- Google Sheets
- Gmail
- Apps Script
- Dunning
Demo build · accounts payable
AI-assisted three-way match for accounts payable
87%invoices matched with no touch
Invoices arriving in the AP mailbox are extracted by a document model, then matched line by line to purchase orders and goods receipts within price and quantity tolerances. Clean matches go straight to posting; exceptions are routed to the buyer with the discrepancy spelled out.
Under the hood
- Line-level PO / GRN / invoice matching with tolerances
- Document-model extraction with confidence gating
- Exception routing to buyer with discrepancy detail
- Power Automate
- AI Builder
- Excel
- Document AI
- AP
Demo build · accounts receivable
AI cash-application engine for accounts receivable
92%receipts auto-applied
Remittance advices in emails, PDFs and spreadsheets are parsed by an LLM into invoice references and amounts, matched to bank receipts and open items, and short-pays are split into deductions with reason codes instead of sitting unapplied.
Under the hood
- LLM remittance parsing across email, PDF and XLSX
- Many-to-many receipt ↔ invoice matching
- Short-pay and deduction coding for follow-up
- Python
- LLM
- Excel
- Cash application
Demo build · expenses
AI expense-policy compliance checker
1,240claims checked monthly
Every claim's receipt is read by a vision model and compared to the claimed amount, date and category, then tested against the policy rules (per-diem caps, hotel limits by city, duplicate submissions, weekend spend) before it reaches an approver.
Under the hood
- Receipt-vs-claim reconciliation with a vision model
- Policy rules engine with city-based caps
- Duplicate detection across claimants and periods
- Python
- LLM
- Google Sheets
- Vision AI
- Compliance
Demo build · payroll
Employee-level benefits and pension reconciliation
0unexplained variances at close
Payroll deduction and employer-contribution reports are reconciled to each provider's invoice per employee and per scheme; joiners, leavers, salary changes and missed enrolments are identified as the cause of every difference.
Under the hood
- Employee-level matching across payroll and provider files
- Root-cause tagging: joiner, leaver, rate change, missing
- Monthly variance report with carry-forward of open items
- Excel
- Power Query
- Payroll exports
- Payroll
- Reconciliation
Demo build · procurement
Supplier onboarding portal with bank-detail verification controls
0unverified bank-detail changes
Vendors self-register through a portal; tax IDs and documents are validated, any bank-detail change triggers call-back verification and dual approval, and only fully approved records are exported to the ERP vendor master.
Under the hood
- Segregation of duties: requester ≠ approver ≠ verifier
- Bank-detail change control with call-back step
- Approved-only export to the ERP vendor master
- Web app
- Supabase
- Approvals
- Internal controls
- RBAC
Demo build · inventory accounting
FIFO inventory-costing and COGS engine
48kmovements costed per month
48,000 monthly stock movements are processed into FIFO cost layers per SKU and warehouse; issues consume layers in order, landed costs are capitalised, and the closing valuation and COGS tie to the general ledger.
Under the hood
- Layer consumption algorithm per SKU and location
- Landed-cost capitalisation onto receipts
- Valuation-to-GL tie-out and negative-stock alerts
- Excel
- Power Query
- Data modeling
- Inventory accounting
Demo build · e-commerce
Unit-economics engine for marketplace sellers
Fees, shipping, ad spend and COGS are applied per SKU so true contribution margin is known before you list.
Under the hood
- Fee schedules by marketplace
- Contribution-margin per SKU
- What-if pricing scenarios
- Google Sheets
- Formulas
- Unit economics
Demo build · insights
LLM-powered survey analytics inside Google Sheets
Hundreds of open-text answers are clustered into themes with sentiment and representative quotes, all inside the sheet.
Under the hood
- LLM theme clustering
- Sentiment scoring
- Quote extraction per theme
- Google Sheets
- GPT
- Apps Script
- NLP
Demo build · customer service
AI ticket triage and routing
Incoming support tickets are classified by topic and urgency by an LLM and assigned to the right queue automatically.
Under the hood
- Topic + urgency classification
- Routing rules by skill
- Helpdesk API write-back
- Python
- LLM
- Helpdesk API
- Classification
Demo build · legal ops
Contract intelligence: clause and term extraction
Parties, dates, renewal terms, notice periods and amounts are extracted from contract PDFs into a structured review sheet.
Under the hood
- LLM extraction with JSON schema
- Clause location references
- Renewal-date alerts
- Python
- LLM
- Document AI
Demo build · reporting
Automated KPI digest with anomaly detection
Yesterday's KPIs, trends and statistical anomalies are delivered to the team's inbox every morning at 8:00.
Under the hood
- Metric computation layer
- Anomaly thresholds vs. baseline
- HTML email rendering
- Apps Script
- Gmail
- Google Sheets
- Anomaly detection
Demo build · operations
Relational re-architecture of a 40-tab workbook
A sprawling workbook is remodelled into linked Airtable tables, views, forms and automations that the team can't break.
Under the hood
- Relational data model design
- Interfaces and permissioned views
- Native automations
- Airtable
- Automations
- Data architecture
Demo build · data
Governed SQL semantic layer for SMB reporting
Clean, documented SQL views on top of operational data give every dashboard one trusted definition of each metric.
Under the hood
- Semantic views per business domain
- Metric definitions as code
- Access grants per tool
- Postgres
- SQL
- Looker Studio
- Data governance
Demo build · VBA
Validated data-entry application inside Excel
A VBA UserForm application with validation, dependent dropdowns and duplicate checks keeps the data clean from day one.
Under the hood
- UserForm UI with dependent lists
- Validation and duplicate checks
- Protected data tables
- Excel
- VBA
- UserForm
- UX
Demo build · VBA
Excel-to-PowerPoint board-pack generator in VBA
24slides, one click
A VBA engine drives PowerPoint through late-bound automation: it opens the corporate template, pastes charts and tables into named placeholders, updates KPI text boxes and commentary, and saves a dated PDF and PPTX, so the board pack is rebuilt in one run.
Under the hood
- PowerPoint object model driven from Excel VBA
- Named-placeholder mapping, no hard-coded coordinates
- PDF + PPTX export with version stamping
- Excel
- VBA
- PowerPoint
- Office automation
Demo build · VBA performance
Performance re-engineering of a legacy VBA codebase
40m → 90smacro runtime
A macro that looped cell by cell, selected sheets and recalculated constantly was rewritten around in-memory arrays, dictionaries for lookups and controlled calculation, then regression-tested to produce byte-identical output in a fraction of the time.
Under the hood
- Range ↔ Variant-array I/O instead of cell loops
- Scripting.Dictionary lookups replacing nested loops
- Output regression test against the legacy version
- Excel
- VBA
- Performance
- Refactoring
Demo build · reporting
Report-bursting engine for 120 branch P&L workbooks
120branch packs per run
A Python engine slices the consolidated ledger by branch and writes 120 fully formatted Excel workbooks with live formulas, conditional formatting and charts, then distributes each one to the right manager's folder with access limited to them.
Under the hood
- Template-driven openpyxl generation with live formulas
- Branch-level data slicing from one model
- Permissioned distribution and run manifest
- Python
- openpyxl
- Excel
- Report bursting
Demo build · finance data
Incremental ELT from an accounting API into a finance data mart
nightlyfinance mart refresh
Journals, invoices, payments and contacts are pulled incrementally from the Xero API with OAuth token rotation and rate-limit handling, landed in Postgres and modelled into fact and dimension tables that Power BI reads directly.
Under the hood
- Incremental loads by modified-since watermark
- OAuth 2.0 token rotation and rate-limit backoff
- Star schema: GL fact, account, contact, date dims
- Python
- Xero API
- Postgres
- ELT
- Data warehouse
Demo build · SaaS analytics
Subscription-metrics engine built on raw billing events
6SaaS metrics, one definition
Subscription events from the billing API are replayed into a daily MRR ledger per customer, producing an MRR bridge (new, expansion, contraction, churn, reactivation), cohort retention curves and NRR/GRR that finance and the board read from one definition.
Under the hood
- Event-sourced daily MRR ledger per customer
- MRR bridge with five movement categories
- Cohort retention and NRR / GRR calculations in SQL
- Python
- Stripe API
- Looker Studio
- SaaS metrics
- SQL
Demo build · manufacturing
Shift-level OEE and downtime analytics for a production line
78%OEE, up from 64%
Operators log stops, output and scrap per shift on a tablet form; the model computes availability, performance and quality against ideal cycle times, Pareto-ranks downtime reasons and alerts the supervisor when a shift falls below target.
Under the hood
- OEE decomposition against ideal cycle time per SKU
- Downtime reason Pareto by line and shift
- Below-target alerts pushed during the shift
- Google Sheets
- Apps Script
- Looker Studio
- Manufacturing
- KPI design
Demo build · internal tool
Cost-centre budget-holder portal with variance write-back
60budget holders self-serving
A small web app gives each of 60 budget holders a secure view of actuals, open POs and forecast for their cost centres only; variance comments are written back to the database and flow straight into the monthly management pack.
Under the hood
- Row-level security scoped by cost-centre ownership
- Actuals + commitments + forecast in one view
- Comment write-back consumed by the reporting pack
- Web app
- Supabase
- SQL
- RLS
- Full-stack
Demo build · S&OP
Statistical demand-forecasting baseline for S&OP
−31%forecast error vs. manual
Sales history for 2,400 SKUs is cleaned of stock-out distortion and fitted with ETS and seasonal-naive models chosen per SKU by back-testing; planners override in Excel, and the consensus forecast flows into purchasing and the revenue plan.
Under the hood
- Per-SKU model selection by rolling back-test
- Stock-out censoring and promo flagging
- Forecast value-added tracking of manual overrides
- Python
- statsmodels
- Excel
- Forecasting
- Statistics
Demo build · data migration
ERP master-data cleansing and de-duplication pipeline
22k → 14krecords after de-duplication
Customer and supplier masters from two legacy systems are standardised, blocked by key attributes and fuzzy-matched into survivor records with a full merge map, so open items and history can be re-pointed during the ERP migration.
Under the hood
- Blocking + fuzzy scoring on name, tax ID and address
- Survivorship rules and a full old→new merge map
- Validation report signed off before load
- Python
- Fuzzy matching
- Excel
- Data migration
Demo build · bookkeeping
Hybrid rules + AI transaction categorisation
Bank transactions are categorised by deterministic rules first and an LLM for the leftovers, with every decision explainable.
Under the hood
- Rules engine first pass
- LLM fallback with reasons
- Accountant-ready export
- Python
- LLM
- Excel
- Hybrid AI
Demo build · services
Field-service job profitability analytics
Labour, parts and travel costs are rolled up per job and compared with revenue, so unprofitable jobs surface automatically.
Under the hood
- Cost allocation per job
- Margin analysis by tech and type
- Below-target alerts
- Google Sheets
- Looker Studio
- Profitability
Demo build · audit readiness
Audit PBC request tracker with evidence-folder automation
184auditor requests tracked
The auditors' prepared-by-client list is loaded into a tracker that creates a Drive evidence folder per request, assigns owners, sends escalating reminders and gives the audit team a shared, read-only status view until every item is accepted.
Under the hood
- Folder-per-request generation with scoped sharing
- Owner reminders escalating to the controller
- Read-only external status view for the auditors
- Google Sheets
- Apps Script
- Drive API
- Audit
Demo build · workforce scheduling
Constraint-aware shift-rota builder with labour-cost guardrails
42staff scheduled weekly
The rota checks every assignment against availability, contracted and maximum hours, rest-period rules and required cover per slot, while live labour cost (including premiums for nights and weekends) is compared with the weekly budget.
Under the hood
- Constraint checks: availability, rest periods, max hours
- Cover-requirement gaps flagged per slot
- Labour cost vs. budget, including shift premiums
- Google Sheets
- Apps Script
- Calendar
- Scheduling
Demo build · nonprofit finance
Restricted-fund and grant budget-tracking system
11restricted funds, one view
Expenses are coded to funds and budget lines, eligibility rules per grant are enforced, and burn rate is projected against the grant period so overspend and underspend risks are flagged before the funder's reporting deadline.
Under the hood
- Fund accounting with per-grant eligibility rules
- Burn-rate projection to grant end date
- Funder report templates generated per period
- Google Sheets
- Apps Script
- Looker Studio
- Fund accounting
Demo build · operations
Inspection-to-report pipeline with corrective-action tracking
< 1 mininspection to filed report
Each submitted inspection is scored with weighted criteria, rendered into a branded PDF with photos, filed in the site's Drive folder and emailed; every failed item becomes an action with an owner and deadline that escalates until it is closed.
Under the hood
- Weighted scoring model per inspection type
- PDF rendering with embedded photos via Docs templates
- Corrective-action log with escalation triggers
- Google Forms
- Apps Script
- Docs
- Workflow
Demo build · real estate
Comparable-sales scraper and automated valuation model
Comparable sales around an address are collected, filtered by similarity and turned into a price-per-m² valuation.
Under the hood
- Geo-radius comp search
- Similarity filtering
- Valuation model output
- Python
- Scraping
- Google Sheets
- Valuation
Demo build · finance
Multi-currency receivables with live FX conversion
Invoices in any currency are converted at live rates, with realised and unrealised FX tracked in your reporting currency.
Under the hood
- Live FX rate feeds
- Realised vs. unrealised FX
- Base-currency reporting
- Google Sheets
- GOOGLEFINANCE
- FX
Demo build · group finance
Multi-entity consolidation with intercompany eliminations
Six legal entities, three currencies, one group P&L. Trial balances are mapped, translated, intercompany eliminated and consolidated on refresh.
Under the hood
- Chart-of-accounts mapping per entity
- FX translation (closing / average)
- Automated intercompany eliminations
- Excel
- Power Query
- Consolidation
- FX
B2B data client
18,000-company enrichment engine with LLM verification
18,000companies enriched
A multi-stage pipeline resolves each company's real domain, finds decision-makers through search APIs, and uses an LLM to verify matches before anything reaches the outreach list.
Under the hood
- Waterfall enrichment across several data sources
- LLM-based entity matching to kill false positives
- Rate-limited async workers with checkpointing
- Python
- LLM
- Search APIs
- Async
All Points North
AI-classified directory built from unstructured public data
100%built without manual entry
Public sources are crawled, parsed, de-duplicated and classified by an LLM into a clean, categorised member directory. No one typed a single row.
Under the hood
- Multi-source crawl with schema normalisation
- LLM taxonomy classification
- Fuzzy de-duplication across sources
- Python
- Scraping
- AI classification
- NLP
Real-estate investor · US
Automated FSBO lead pipeline into the CRM
★ 5.05.0 ★ client review
For-sale-by-owner listings are collected every day, de-duplicated against existing records, enriched and pushed into the CRM before the acquisitions team logs in.
Under the hood
- Daily scheduled extraction
- Dedup against CRM by address + phone
- CRM API upload with field mapping
- Python
- Scraping
- CRM
- CRM API
Healthcare directory · US
Human-in-the-loop AI cleanup of a healthcare directory
An LLM normalises names, credentials, phones and specialties across thousands of listings; anything below a confidence threshold goes to manual enrichment.
Under the hood
- LLM normalisation with strict output schema
- Confidence threshold routing
- Manual QA on flagged records
- Python
- LLM
- Manual QA
- Human-in-the-loop
Web3 marketing client
Web3 market map: game studios and their community channels
★ 5.04.5 ★ client review
Blockchain game projects are identified and enriched with Discord, X and email channels, giving a marketing team a ready-made outreach universe.
Under the hood
- Multi-platform identity resolution
- Social link discovery and validation
- Outreach-ready export
- Python
- Scraping
- Enrichment
- Identity resolution
Lending & lead-gen client · US
Multi-state UCC filings intelligence (FL · CA · CO)
Secured-transaction filings across three state registries are searched, extracted and normalised into one dataset, so lenders can spot financing activity across borders.
Under the hood
- Three registry formats unified
- Debtor-name normalisation
- Secured-party flagging
- Python
- Public records
- Data extraction
Data client · US
Georgia public-records extraction with per-record cost model
Records from Georgia's clerk authority are extracted county by county into structured data, with a transparent per-record pricing model for the client.
Under the hood
- County-level extraction workflows
- Record-type classification
- Unit-economics costing model
- Python
- Scraping
- Public records
- Costing
Legal services client
Weekly New Jersey court-filings intelligence feed
New court filings are collected every week, structured by county and type, and delivered as a subscription report to paying subscribers.
Under the hood
- Weekly scheduled collection
- County and case-type structuring
- Automated subscriber delivery
- Python
- Scraping
- Subscriptions
B2B data client · Switzerland
Swiss commercial-registry (Zefix) data extraction at scale
Company records are extracted from Switzerland's official registry, cleaned and enriched with domains, producing a targeting dataset by canton and status.
Under the hood
- Registry API / page extraction
- Field cleanup across 4 languages
- Domain enrichment and validation
- Python
- Scraping
- Zefix
- Multilingual data
Construction lead-gen · France
Certified-contractor intelligence app (Qualibat · OPQIBI)
France's certified-contractor directories are scraped and turned into a searchable application for construction lead generation.
Under the hood
- Two directories, one unified schema
- Scraper-to-app data pipeline
- Search and filter interface
- Python
- Scraping
- Web app
- Full-stack
Food & beverage supplier
Local-business enrichment for a B2B sales team
A list of hospitality businesses is enriched with websites, social profiles, contacts and size signals, ready for segmented outreach.
Under the hood
- Multi-source web enrichment
- Social profile discovery
- Segmentation-ready output
- Python
- Enrichment
- Google Sheets
- Lead data
Demo build · lead data
Web-scale lead discovery with Common Crawl
Billions of crawled pages are queried to find the domains in a niche that carry a contact form: lead discovery at internet scale.
Under the hood
- Common Crawl index querying
- Form detection heuristics
- Domain-level de-duplication
- Python
- Common Crawl
- Big data
I take on 4 new projects a month · 2 spots left for this month
Questions people ask before the call
Do we have to change the tools we use?
No. Most of what I build lives inside the tools you already pay for: Google Sheets and Workspace, Excel and Microsoft 365, your CRM. When a small web app is the better answer, it plugs into those same files.
How much does it cost?
It depends on the process, so I don't guess on a website. After the call you get a written plan with one fixed price, before any work starts. Small fixes can take a few hours; full apps take a few weeks.
Who owns what you build?
You do. The code, the files and the accounts are in your name, and you get a written handover so anyone on your team can run it.
What happens when something changes later?
You message me. I've kept one client's weekly automation running for four years, fixing it whenever something upstream changed.
Is our data safe?
Automations run inside your own accounts and servers wherever possible, with the minimum access needed. Happy to sign your NDA.
Lateef Moradee · I read every request myself
Send me the task your team still does by hand.
In 20 minutes I'll tell you if it's worth automating and what it would take. If it isn't worth it, I'll say so.
Get my free automation plan
2-minute formFree 20-min callFixed quote, no obligation
Already know what you need? Book the call directly