// software engineer · ai systems

Muhammad Kashif-Khan

I take on problems I've never seen and ship working software for them: biomedical simulation, scientific tools, computer vision, cloud and AI. MEng Computer Science & AI, University of Leeds, First-Class Honours.

Open to graduate software & AI roles

// tech stack

What I work with

// languages
  • Python
  • C#
  • C/C++
  • TypeScript
  • SQL
// ai & vision
  • PyTorch
  • OpenCV
  • scikit-learn
  • LLM agents
  • MCP
// web & cloud
  • ASP.NET Core
  • FastAPI
  • React
  • PostgreSQL
  • Azure
// systems & hpc
  • CUDA
  • OpenMP
  • MPI
  • Docker
  • Linux

// selected work

Five projects, five different fields

Each one was new to me when I started.

// biomedical

Flow Diverter Deployment & Medical Simulation Platform

Built a GPU-accelerated tool for simulating flow-diverter stents in blood vessels, with stent models and joint-level expansion forces.

Python 3.10VMTKPyVistaPyQt6
Read the case study →

// scientific software

pNanoLocz: Python AFM Analysis Platform Development

Led the rebuild of a MATLAB microscopy tool in Python, making it about 50% faster, and mentored an intern.

PythonPyQt6NumPyMATLAB
Read the case study →

// computer vision

3D Reconstruction from a Single Video

Led a team of five building a 3D reconstruction pipeline from phone video, with 1.56 cm camera error on Replica.

PythonPyTorchOpen3DComputer Vision
Read the case study →

// cloud

Azure Serverless Inventory Management System

Built an event-driven serverless inventory system on Azure Functions and Cosmos DB, benchmarked for cold starts and throughput.

AzureServerlessDistributed SystemsCloud Computing
Read the case study →

// infrastructure

Home Lab & Remote Infrastructure

Set up local LLMs and coding agents on a home GPU server, reachable anywhere over an encrypted Tailscale mesh.

TailscaleLinuxllama.cppDocker
Read the case study →
All projects → Robotics, graphics, operating systems, parallel computing and more.

// what I'm building

Built in my own time

Two projects I keep coming back to, both about developer tools and AI agents.

Synergy

in development

Turns a codebase into a graph you can query: what calls what, what a change will break, how two versions differ. It runs on your own machine, and coding agents can use it over MCP.

PythonReactTypeScriptSQLitetree-sitterMCP
Where it is now →

HireForge

in development

Coding assessments run by AI agents. One agent builds a small codebase with planted bugs, the candidate fixes them in a sandbox with an AI assistant, and a separate agent reviews their work and writes the scorecard.

PythonFastAPIPostgreSQLReactTypeScriptDocker
Where it is now →

// experience

Experience & education

  1. 2022 – 2026

    MEng Computer Science with AI

    University of Leeds

    First-Class Honours. I led my master's group project, a pipeline that reconstructs 3D scenes from a single video.

  2. Jun – Aug 2025

    Summer internship: stent deployment simulation

    University of Leeds

    Built a GPU-accelerated tool for simulating flow-diverter stents in blood vessels, with stent models and joint-level expansion forces.

  3. Jun – Aug 2024

    Summer internship: pNanoLocz

    University of Leeds

    Rebuilt a MATLAB microscopy analysis tool in Python, about 50% faster than the original. I led the team and mentored a junior intern.

  4. At university

    Core Executive Leadership Team

    Leeds Artificial Intelligence Society

    Organised and led technical workshops and helped run the society.

// contact

Let's talk

Interested in collaborating or discussing opportunities? I'd love to hear from you. Fill out the form below and I'll get back to you as soon as possible.

Let's Connect

Response time

I typically respond to inquiries within 24-48 hours. For urgent matters, please mention it in your message.

Ask AI

answers from this site · DeepSeek

Starting assistant...