Open to Freelance & Full-Time Opportunities
AI Application Engineer
I Turn Ideas Into Intelligent Applications.
I design, build and ship AI-powered applications, developer tools and complete software experiences across web, desktop and Android.
- 22Projects documented
- 14AI applications and experiments
- 10With public repositories
- Python
- FastAPI
- React
- TypeScript
- RAG
- LLM Apps
- ChromaDB
- PyTorch
- Kotlin
- Jetpack Compose
- PySide6
- SQLite
- Next.js
- Supabase
- Tailwind CSS
- Ollama
- OpenCV
- Git & GitHub
About
Software that works, not just prototypes.
I'm Muhammad Umar Jamal, a BSCS student and AI application engineer based in Okara, Punjab. I focus on turning ideas into real, usable software.
That means pairing AI capabilities such as language models, retrieval, vision and automation with the engineering around them: backends, interfaces, packaging and deployment.
My work spans AI assistants, document intelligence, computer-vision systems, Android apps, desktop software and low-level systems programming. I lead Group #2 on university projects, and I learn fastest by shipping.
- Name
- Muhammad Umar Jamal
- Brand
- MUdevelops
- Location
- Okara, Punjab, Pakistan
- Education
- BSCS, University of Central Punjab, Okara Branch (2023–2027)
- University role
- Group lead, Group #2
- Instantregret96@gmail.com
- Phone
- +92 321-9211464
- GitHub
- github.com/MUdevelops
AI Engineering
What I can build with AI
Eleven capabilities, each tied to the projects where I've applied it. Use the project buttons to open the case study.
These come from personal and university projects. Some projects are still in progress or experimental, and they are labelled that way.
RAG systems
Computer vision
AI assistants
AI-powered applications
The layers behind most of my systems
Projects
Work I've built and shipped
AI applications lead, supported by web, desktop, Android and systems engineering. Every card shows its real status, and repository links go straight to the project.
Skills
Tools I work with
Grouped by where they sit in a system. No percentage bars: the projects above are the evidence.
Languages
Application and systems code.
AI and ML
Models and the pipelines around them.
Backend and data
Services and storage.
Frontend
Interfaces people use.
Mobile and desktop
Apps that install and run locally.
Tools and systems
Version control, build and networking.
Engineering process
How I build AI applications
Architecture, model integration, testing, interface quality and deployment are decided together, not bolted on at the end.
- 1
Understand
Define the real problem and what “working” means before touching a model.
- 2
Design
Decide what the model does and what the application around it handles.
- 3
Build
Connect models, APIs and retrieval to a real backend, data store and interface.
- 4
Test
Check outputs and behavior on real inputs, not only happy-path demos.
- 5
Ship
Package, document and deploy so someone other than me can use it.
- 6
Improve
Use feedback and failures to tighten the next version.
GitHub
The code, in the open
Source for the projects below lives on GitHub. This list works without any network call; live details are added when GitHub responds.
github.com/MUdevelops
AI applications, developer tools, Android, desktop and systems projects.
Checking GitHub for live repository details…
Contact
Let's build something.
I usually reply within a day. Tell me what you're making and where it needs to run.
- Freelance software projects
- AI application development
- Internships and relevant employment
- Technical collaborations
- Project inquiries