Backend / systems engineer · Lahore, open to remote roles
I build the systems behind messy, multi-system workflows, and then I keep them running. My habit is to find the problem, understand the system underneath it, build the fix, and stay for what breaks next.
Mubashir RehmanLahore · UTC+5
At TransData, where I have led a team of five engineers since March 2026, I took an AI voice front desk for dental clinics from AI-scaffolded code to a production launch, then designed and deployed its AWS backend on services verified as HIPAA-eligible. I have also built agents that work inside a live ERP, and a social-signal intelligence backend I wrote alone for a client.
Before that, at VeritusLabs, I was promoted to team lead about three months in and led four engineers on drone-control software. For five semesters I was a teaching assistant for operating systems at Information Technology University.
All ten rules. Each one came out of real work, and the link is the evidence.
Rule 1: Read the number that does not add up.
Rule 2: Split diagnosis from fixing where the blast radius changes.
Rule 3: Prove the path end to end before you tune the thresholds.
Rule 4: Keep the adapter generic. Put business rules where the business trigger lives.
Rule 5: A partial result should look partial.
From An operations console for a healthcare integration engine →
Rule 6: Check the vendor's own list, not your memory.
Rule 7: Write the specification first, and trace every requirement to a screen.
From Architecture for a multi-tenant lead-intelligence SaaS →
Rule 8: Deterministic first, model second.
From HireTrack: an AI résumé tailor and job-application tracker →
Rule 9: Report a result with its sample, or do not report it.
Rule 10: Give the agent today's docs, not its memory.
Full-time roles count toward the 3+ years; teaching and the internship do not.
Leads five engineers: one AI engineer and four full-stack engineers. Backend, AI and infrastructure across healthcare, ERP and AI products. Now: an AI voice front desk for dental clinics and its AWS backend.
Full-time
Backend, AI and infrastructure across healthcare, ERP and AI products.
Full-time
Promoted about three months in. Led four engineers on drone-control software in C++ and Qt/QML. Built an AI prompt-management app and scripted AWS provisioning.
Full-time
Drone ground-control features in C++ and Qt/QML.
Full-time
Operating systems: undergraduate labs for three semesters (50+ students), graduate Advanced Operating Systems for two (30+ students).
Part-time, alongside full-time work
Android and web endless-runner games in Unity and PlayCanvas.
Internship
Explaining systems, and measuring them.
For five semesters I was a teaching assistant for operating systems at Information Technology University: I ran undergraduate labs on file systems, threads, scheduling, system calls and synchronisation, discussed classic systems papers with graduate students, wrote assignments under the instructor and graded projects.
Research · Circuits, Systems, and Signal Processing (Springer), 44(8), 5852 to 5875, 2025
Co-author with equal contribution, third of seven authors. My part: data collection, the signal-processing and feature pipeline, and model training and evaluation. 42 subjects and 6,131 segmented samples; a random forest reached R² 0.99 on segmented data and R² 0.87 with transfer learning on unsegmented data.
DOI 10.1007/s00034-025-03048-2arXiv preprintHow the study was built
Where it started, and what I use now.
Information Technology University, Lahore.
The full toolbox, from daily tools to project exposure, is on the home page.