Started BS Computer Science
Enrolled at Bahria University, Karachi Campus — pursuing a four-year CS degree.
Brew Haven
Full-stack e-commerce coffee shop — Express.js, Node.js, JWT auth, RBAC, React, MongoDB
Available for freelance
Building systems that think and interfaces that ship
I'm Aun — a Computer Science student at Bahria University, Karachi. I work across the full stack: training neural networks, shipping REST APIs, and managing development teams end-to-end.
About Me
CS student, ML researcher, and full-stack developer — based in Karachi
Hello there
Hi, I'm Aun — a CS student building things that think and things that ship
I work across the full spectrum: training neural networks on real-world datasets, shipping REST APIs and React frontends, and managing development teams from planning through delivery.
I'm currently doing independent ML research under a faculty supervisor — building physics-informed neural networks and epidemiological models in Python. On the product side, I lead multi-track projects for a stealth startup. I like my code clean, my models validated, and my timelines realistic.
Machine Learning
End-to-end ML pipelines on real-world datasets — trained, validated, visualized.
Full-Stack Web
Express/Node.js backends, JWT auth, REST APIs, and React frontends.
Data Analysis
Statistical modeling and visualization with Python, R, Power BI, and Seaborn.
Project Management
Sprint planning, WBS, Gantt charts, and distributed team coordination.
IT Certification
Completed an IT foundation certification at the Global Institute.
Research Developer
Independent ML research under Dr. Oyoon Abdul Razzaq — fPINNs and epidemiological modeling.
Project Manager
Leading delivery across two concurrent product tracks at a stealth startup.
"I like my code clean, my models validated, and my timelines realistic."
Clean Code
Lo-fi Beats
Karachi
Problem Solver
Skills
From neural networks to REST APIs — tools and technologies I work with
Machine Learning
AdvancedEnd-to-end ML pipelines — LSTM, CNNs, fPINNs — trained and validated on real-world datasets with full visualization output.
Deep Learning
AdvancedPyTorch-based architectures for image classification, time-series prediction, and solving PDEs with physics-informed networks.
Full-Stack Web Dev
StrongExpress.js and Node.js backends with JWT auth, role-based access control, REST APIs, React frontends, and MongoDB persistence.
Data Analysis
StrongStatistical modeling, data cleaning, and visual storytelling using Python, R, Power BI, Matplotlib, and Seaborn.
Python & JavaScript
ProficientPrimary languages for ML research and web engineering — also comfortable with C++, R, and SQL.
Project Management
ExperiencedSprint planning, WBS, risk registers, Gantt-based timelines, and stakeholder alignment across distributed teams.
Toolbox
- Python, JavaScript, C++, R, SQL
- React, Node.js, Express.js, MongoDB
- PyTorch, Matplotlib, Seaborn, Power BI
- Git, ClickUp, Instagantt
I build across domains — from physics-informed neural networks trained over 4,000 epochs, to full-stack coffee shop storefronts with JWT-secured APIs. The throughline is the same: understand the problem, pick the right tools, and ship something that works.
Resume
Experience, education, and core competencies at a glance
Professional Experience
Hands-on roles spanning ML research, full-stack development, and project delivery.
Project Manager
Stealth Startup
2025 – Present
Led delivery across two concurrent product tracks — defined scope, assigned ownership, and ran weekly syncs to surface blockers early. Maintained Gantt-based sprint plans in ClickUp and Instagantt, keeping a distributed team of six aligned on milestones and delivery timelines.
Research Developer — Computational Modeling
Independent Research under Dr. Oyoon Abdul Razzaq
2025 – Present
Implemented a SEIR epidemiological model with seasonal forcing and nonlinear incidence rates in Python, producing animated visualizations demonstrating chaotic divergence and the butterfly effect. Built a fractional Physics-Informed Neural Network (fPINN) in PyTorch to solve PDEs via the fractional Laplace transform — trained over 4,000 epochs and validated against exact analytical solutions with 2D, 3D, and animated output rendering.
Portfolio
A selection of projects across web engineering and AI/ML
- All
- Web Engineering
- AI / ML
Stock Price Prediction
Stacked LSTM on OHLC data with MinMaxScaler · evaluated with RMSE & MAE
Breast Cancer Classification
Hybrid CNN fusing mammographic & sonographic features — Final Year Project (In Progress)
Computational Modeling
Neural networks + epidemiological chaos · PyTorch + SciPy implementation
Testimonials
What collaborators and clients have to say
Services
What I build and how I can help your project move forward
Full-Stack Web Development
Custom web applications built with React, Node.js, and Express — from REST API design and authentication to frontend deployment.
AI & ML Solutions
End-to-end machine learning pipelines: data preprocessing, model training, evaluation, and visualization — tailored to your dataset and goals.
Personalized AI Chatbots
Intelligent, context-aware chatbots tailored to your product, workflow, or customer support needs — built and integrated end-to-end.
Upscaling & Optimization
Performance audits, SEO improvements, and modernization of existing sites — bringing legacy web properties up to current standards.
Maintenance & Support
Ongoing support to keep your web applications running smoothly, securely, and up to date — with fast turnaround on fixes and updates.
Technical Consulting
Architecture planning, tech stack decisions, and project scoping for early-stage startups and teams — clarity before you commit resources.
Frequently Asked Questions
Straight answers to the questions I get most often
01
What does your development process look like from brief to delivery?
I start by understanding what you actually need — not the feature list, but the goal. From there I scope the project, build the backend and API first, then the frontend, followed by testing and deployment. You get a clear timeline upfront and weekly check-ins throughout. Nothing goes dark.
02
Do you handle both the backend and frontend, or just one side?
Both. I work full-stack — Express.js and Node.js on the backend, React on the frontend, MongoDB for persistence. I design the API, implement authentication, build the UI, and wire it all together. You get one person responsible for the whole thing.
03
How do you approach ML projects — what does the pipeline look like?
Every ML project starts with a problem definition and a data audit before any model is touched. I then handle preprocessing, architecture selection, training, and evaluation — with validation metrics tracked throughout. Deliverables include the trained model, clean code, and visualization outputs.
04
I have an existing codebase — can you work with what's already there?
Yes. I review the existing codebase before committing to anything — architecture, dependencies, known issues. If I take it on, I take it on properly. Whether it's a bug fix, a new feature, or an ongoing maintenance retainer, I work with what's there rather than rewriting from scratch.
05
What does availability and turnaround look like for freelance work?
I respond within 24 hours and am available Monday through Saturday. For project work, I'll give you a realistic timeline upfront and stick to it. For maintenance and support, turnaround on reported issues is typically within 24 hours.
Contact
Get in touch — I'll respond within 24 hours
Have a project in mind?
Let's build it.
Use the form below to tell me about your project. I'll get back to you within 24 hours with a response and next steps.