Olutoye A. Odufowokan

Computer Science @ UT Dallas '27. Full-stack web, systems programming, and local AI.

GitHubLinkedInRésuméEmail

Selected work

01

ShopOpen

Open-source e-commerce storefront template

A full-stack, deployable storefront for browsing products, filtering inventory, and managing a cart — packaged so anyone can fork it and launch a shop.

  • Dynamic catalog filtering, instant search, and cart checkout
  • Containerized services deployed to AWS (EC2, S3) with GitHub Actions CI/CD
  • Normalized, indexed PostgreSQL schema to keep product queries fast
  • React
  • TypeScript
  • Node.js
  • PostgreSQL
  • AWS
  • Docker
View source

02

DocGen

Local AI documentation & README generator

Drop in a codebase and get a structured README or API reference back in seconds — generated by a model running on your own machine, so the code never leaves your network.

  • Offline inference through Ollama — no paid cloud APIs, no per-token cost
  • ZIP upload or paste, model selector, and persistent generation history
  • One-command setup with Docker Compose across operating systems
  • React
  • Node.js
  • PostgreSQL
  • Ollama
  • Docker
View source
Encryption daemon architecture: a Go watcher sends new files to a C++ libsodium engine and writes encrypted .enc files

03

Encryption Daemon

Concurrent secure file pipeline

A Linux daemon that watches folders and encrypts incoming files the moment they land, pairing Go's concurrency with a native C++ cryptography engine.

  • Go goroutines and inotify watch directories without polling
  • Authenticated encryption in a C++ engine built on libsodium
  • Cross-language IPC so it runs unattended as a background daemon
  • Go
  • C++
  • libsodium
  • Linux IPC
View source

Building now

In development10 features on the roadmap

Anansi

A local agent harness built to make small models genuinely useful — on the laptop you already own, with no subscription and no API bill.

Read the vision & roadmap

About

I'm a Computer Science student at UT Dallas, graduating in 2027. I like building software that's efficient and stays under the user's control — full-stack web apps, systems tools in Go and C++, and AI that runs locally instead of on someone else's server.

Lately I've been focused on making small models genuinely useful on ordinary hardware. That's the idea behind Anansi, the agent harness I'm building now.

Skills

Languages
TypeScript, JavaScript, Go, C++, Java, C, SQL, HTML5, CSS3
Web
React, Node.js, Express, REST APIs, WebSockets
Infrastructure
AWS (EC2, S3), Docker, GitHub Actions, Linux
AI
Ollama, LM Studio, Local LLM integration