Experience & Toolkit

The tools behind the work.

A focused view of the technologies, systems, and automation patterns I use to build reliable software.

Experience.

Recent work across AI evaluation, full-stack systems, and production-minded project delivery.

Resume-backed

AI Reasoning Engineer

Turing

Worked on training and evaluating large language models with a focus on reasoning, in-context learning, and software engineering problem solving.

Designed complex coding and algorithmic tasks with structured reasoning paths to improve model accuracy and interpretability.

Contributed to CL-Bench, a context-learning benchmark with 1,899 tasks and 31,000+ evaluation rubrics.

Built an internal response extraction and defect analysis tool to identify inconsistencies in LLM outputs and improve evaluation reliability.

Software Engineer Fellow

DevWeekends Fellowship

Built full-stack applications while sharpening scalable engineering workflows, algorithms, review habits, and product delivery discipline.

Built and deployed 3+ full-stack applications using the MERN stack and Next.js.

Solved 150+ LeetCode problems across trees, graphs, recursion, dynamic programming, and advanced algorithmic patterns.

Collaborated through peer code reviews, debugging sessions, and architecture discussions to improve engineering quality and speed.

Project experience

Systems I have shipped end to end

A few concrete builds where the work moved from product idea to architecture, APIs, persistence, deployment, and user-facing flows.

Engineered a multi-vendor commerce platform with role-based dashboards and real-time chat for buyers, sellers, and admins.

/Developed 20+ RESTful APIs for authentication, products, orders, refunds, and payments.

/Implemented JWT authentication, SMTP verification, and centralized middleware-based error handling.

/Designed scalable MongoDB schemas and deployed the distributed architecture across Vercel, Render, and dedicated Socket.IO services.

Built a full-stack learning management system with learner and admin dashboards, secure auth, analytics, and course workflows.

/Implemented JWT and OAuth authentication with Google and GitHub.

/Integrated Redis caching for sessions and course data, reducing average load times by approximately 40%.

/Improved API efficiency with RTK Query caching and built analytics dashboards with ReCharts and Tailwind CSS.

Tech Stack.

Technologies I use to build reliable software

Languages

Core building blocks I use across product and systems work.

Java
Java
JavaScript
JavaScript
TypeScript
TypeScript
HTML
HTML
CSS
CSS

Frontend

Interfaces, state, component systems, and polished user flows.

React
React
Next.js
Next.js
Tailwind
Tailwind
Expo Go
Expo Go
Redux
Redux
Material UI
Material UI

Backend & Data

APIs, real-time systems, persistence, and application services.

Node.js
Node.js
Express.js
Express.js
Socket.IO
Socket.IO
Firebase
Firebase
MongoDB
MongoDB
MySQL
MySQL
PostgreSQL
PostgreSQL
Redis
Redis

Infrastructure & Tools

Deployment, delivery, version control, and cloud assets.

AWS
AWS
Docker
Docker
Git
Git
GitHub
GitHub
Vercel
Vercel
Cloudinary
Cloudinary

AI & Automation

LLM-powered product features, retrieval workflows, agents, and model adaptation.

OpenAI SDK
OpenAI SDK
Open-Source Models
Open-Source Models
LLM Integration
LLM Integration
Agentic AI
Agentic AI
RAG's
RAG's
LangChain
LangChain
LangGraph
LangGraph
HuggingFace
HuggingFace
Model Fine-tuning
Model Fine-tuning

- and always learning more

© 2026 Muhammad Ali Khan. All rights reserved.