Muhammad Umar
ERP Engineer
Leads attendance/payroll, finance and compliance engines, and AI-assisted integrations across both platforms.
ThinkDevs is a small, senior engineering team that builds and rescues production Odoo ERP systems. Payroll that's quietly wrong, compliance that's one audit away from a problem, data entry nobody has time for — these are the same handful of failure patterns wherever they show up. We've solved them across two live platforms, five company instances, five countries.
Click through to see exactly how we solved it, for who, and what changed afterward.
“Payroll is off by half a day for a few people, every single month, and nobody can figure out why.”
See how we solved this“We're manually retyping customer forms into the ERP all day — PDFs, scans, the occasional phone photo.”
See how we solved this“Filing our VAT return is a multi-day spreadsheet exercise, and one missed row means it's wrong.”
See how we solved this“Finance has a depreciation schedule and operations has a spreadsheet of who's holding what — and they don't agree.”
See how we solved this“Our biometric attendance data is technically 'real' but practically fiction — duplicate punches everywhere.”
See how we solved this“Changing a single letterhead logo on our invoices means opening a code editor and redeploying.”
See how we solved this“We need dashboards our team can actually build themselves, without a developer for every new chart.”
See how we solved this“We bill flight/project hours by hand, and disputes over 'how many hours was that really' happen constantly.”
See how we solved this“Our database backups exist, but nobody's confident they'd actually work in an emergency — or that the admin console is even locked down.”
See how we solved this“Sales reps burn hours on repetitive CRM admin — writing the same follow-up checklists over and over.”
See how we solved thisTwo different Odoo versions, two different industries, three engineers moving between them — and the same root cause kept recurring: several parts of a system quietly disagreeing about the same fact. Payroll disagreeing with attendance. Tax logic scattered instead of centralized. A security secret stored somewhere a bad actor could also reach.
Our answer is never a fourth opinion. It's collapsing the disagreement into one definition everyone calls.
Every case study here opens with the sentence a client actually brought us. "Payroll is off by half a day" is a better starting point than a requirements document.
The bug is rarely the arithmetic. It's usually several parts of the system quietly holding different opinions about the same fact.
One day classifier. One tax category. One stock move. Everything else reads from it, so there is nothing left to drift.
Put the rule where it can't be walked around, keep the off switch out of reach, and back it with tests — the payroll classifier shipped with 19.
Every case study on this site comes from one of two production systems we build and maintain.
One Odoo 17 codebase, deployed as four separate company instances: MAFA (parent company, industrial engineering, Jeddah), TechTrade24 KSA and TechTrade24 Germany (B2B industrial trading), and Eesint (Pakistan). Maintained continuously since April 2025.
A second Odoo 18 deployment for a Gulf-based charter aviation and logistics group, operating across the UAE, Saudi Arabia, and Bahrain as three companies under one multi-company instance. Built collaboratively across three engineers since October 2025.
Each case study starts with the problem a client brought us and is credited to whoever led the work.
Four parts of the system each decided what counted as a working day. We replaced them with one classifier everything reads from.
Led by Umar · ~28,000 LOC Read the case studyA drag-and-drop builder driven by typed, whitelisted conditions, so every viewer sees only what their access rights allow.
Led by Yaseen · Team built · ~20,800 LOC Read the case studyA vision model extracts registration forms straight from incoming mail. It never writes to the database — an admin approves.
Led by Umar · ~4,350 LOC Read the case studyA small, senior team — every case study is credited to whoever actually led it.
ERP Engineer
Leads attendance/payroll, finance and compliance engines, and AI-assisted integrations across both platforms.
ERP Engineer
Leads the dashboard/reporting platform, security hardening, and infrastructure work across both deployments.
ERP Engineer
Leads CRM and pipeline automation, including the AI-assisted checklist and activity generation tooling.
The case studies told the long way: the problem, the root cause, and the fix.
Half a day, a handful of people, every month — and four parts of the system with four opinions.
Read the noteWhy our security features keep their off switch in the server config file.
Read the noteThe one constraint that made an AI intake pipeline safe to ship.
Read the noteTell us what's broken. Chances are we've already fixed a version of it.