Selected work
A collection of things I've built, problems I've explored, and systems I've worked on. From backend infrastructure and developer platforms to AI experiments and community tools, every project has taught me something beyond the code itself.
The common thread
Understand the problem.
Build the solution.
Engineering, infrastructure, and product thinking.
The project archive
A centralized identity platform for secure Single Sign-On across the AWS LPU ecosystem. Built around OAuth 2.0 and OpenID Connect, with PKCE, application authorization, user profiles, role management, and developer integration tools.
A decoder-only Transformer language model built from scratch in PyTorch. Includes a custom tokenizer, multi-head self-attention, causal masking, autoregressive generation, configurable sampling, and a complete training pipeline.
An AI-powered community operations platform combining a Discord assistant, community knowledge, feedback collection, human escalation, and a web dashboard into one unified operating layer.
A certification exam platform designed around randomized assessments, timed attempts, multiple question types, configurable passing criteria, question pools, scoring, and certification tracking.
A centralized platform supporting campus events, registrations, attendance, organizational workflows, and automated certificate generation for the AWS LPU community.
A Discord automation platform built for community management, with moderation, server utilities, premium functionality, persistent data, background tasks, integrations, and a dedicated web dashboard.
An agentic multi-stop route planner built with Azure AI Foundry that determines an efficient visiting order across multiple destinations.
The next chapter
I'm interested in problems that push me beyond what I already know. That means exploring deeper system design, building more dependable infrastructure, and turning ideas into products people can genuinely rely on.
The goal isn't to collect projects. It's to keep getting better at understanding problems, making engineering decisions, and building things that work in the real world.
Have something interesting in mind?