Muhammad Zeshan

[full-stack engineer]

I'm a full-stack engineer. For the past four years I've built web apps for healthcare clinics, SaaS startups, real-estate investors, online stores, and AI model evaluation. I usually handle the whole thing myself: the database, the API, and the interface people actually use.


Projects

A few things I've built and shipped — most of them end to end.

County Lead Intelligence

County Lead Intelligence

3 months

A daily pipeline that turns scattered county court and recorder filings into owner-resolved, equity-scored leads for real-estate investors.

AI Agent Evaluation

AI Agent Evaluation

contract (ongoing)

Tasks that test whether an AI agent can build a real, rule-enforcing web app — plus the tooling to quality-check them in hours instead of days.

PieCyfer Website

PieCyfer Website

2 months (ongoing)

A rebuilt company site with its own CMS for blog, jobs, and page metadata — every existing page carried over.

Damascus Frontier

Damascus Frontier

3 months

A direct-to-collector store for a blade maker — 71 products, Stripe checkout, and an admin that handles its own media.

Instagram DM Intake

Instagram DM Intake

Turns a burst of Instagram DMs into exactly one AI reply — verified, deduplicated, and handed back to a human when the coach steps in.

Bar Ordering & POS

Bar Ordering & POS

3 months

Online ordering and an offline-capable POS for a neighbourhood bar — tabs, split bills, shifts, and end-of-night reconciliation.

Physician Job Board

Physician Job Board

6 weeks

A legacy AngularJS job board for physicians and facilities, migrated to a modern NestJS and Next.js monorepo.

TestFiesta

TestFiesta

1 year (ongoing)

A test management tool that bends to how QA teams actually work — with an AI Copilot, versioned test history, and flat $10/user pricing.

Millennium Medical

Millennium Medical

2 years (ongoing)

Automated scheduling, reminders, and refill tracking for an ADHD clinic that was running thousands of patients by hand.

Form Maker

Form Maker

5 months

A no-code tool for building forms, invoices, and queryable data tables — one app instead of four.

Fireball

Fireball

1.5 year

A multiplayer game that teaches remote teams Agile by having them actually play it — ceremonies and all.

DI

DistrictCSA

8 months

Real-time tracking for 100+ hospital devices — staff find equipment about 40% faster.

SDASH

SDASH

6 months

A real estate CRM that moved a paper-run operation onto one system — sales, construction, tenancy, and admin.

OnlineDoc

OnlineDoc

10 months

One secure web and mobile app for doctors — tasks, patient communication, and scheduling in a single place.

Atlanta Dog Trainer

Atlanta Dog Trainer

1 year

A new site for a 30-year dog-training business — 400+ five-star reviews, 10,000+ dogs trained, online booking.


Experience

TUTuring
2026 – Present

Contract Engineer, AI Agent Evaluation

  • -Design full-stack web-app tasks that test whether AI coding agents can build real, rule-enforcing apps
  • -Write LLM-judge rubrics that grade server-side enforcement by replaying recorded requests as other users
  • -Calibrated one task from about 98% to about 42% model score while the reference solution stayed at 100%
  • -Built the QC tooling: runtime-faithful launcher, static rule linter, deterministic packaging, and a multi-agent review loop
PIPieCyfer
2022 – Present

Senior Full-Stack Engineer

Skills

ReactNext.jsNode.jsTypeScriptNestJSGraphQLPostgreSQLMongoDBFirebaseSocket.ioPythonPrismaStripen8nWeb ScrapingPlaywrightDockerVercelAILLM EvaluationAWSGoogle Cloud

Get in touch

I'm open to full-time roles, contract work, and consulting. Email is the fastest way to reach me — always happy to talk through what you're building.