Learning sites, developer tools, and systems projects. Private repositories stay unlinked. The sites are public.
28 tracks and 343+ challenges, from message passing to consensus
A hands-on course for distributed systems. You implement real algorithms against Maelstrom-style tests: message passing, gossip, Raft, storage engines, queues, and failure handling, in eight languages. The repository is private. The site is public.
479 tasks in C, assembly, compilers, CPUs, and operating systems
A practice site for low-level engineering. Tracks cover language foundations, compilers, CPU architecture, firmware, operating systems, performance, binary analysis, networking, virtual machines, graphics, and storage. The repository is private. The site is public.
Low-level programming roadmap with 1,356 GitHub stars
A curated low-level programming roadmap and resource hub covering C, x86 assembly, systems programming, deep learning, and neural networks. Forked 89 times for personal learning tracks.
142 agent skills for C, C++, Rust, and Zig. 226 GitHub stars.
A curated suite of AI agent skills for systems and low-level programming, covering compilers, debuggers, profilers, build systems, sanitizers, and binary analysis across C, C++, Rust, and Zig. 51.1k installs on skills.sh, and 226 GitHub stars. Published at lowleveldevskills.com.
Geospatial intelligence engine for tech conferences
A premium conference tracking engine that aggregates CFPs and tech events into a single searchable, geospatial interface so developers stop missing deadlines buried in scattered Twitter threads and wikis.
Terminal dashboard for AI coding token & cost tracking
A zero-dependency terminal dashboard that reads AI coding assistant session data and surfaces token usage, cost, and cache efficiency at a glance. Helps developers understand what their AI assistants actually do with their tokens.
Go Markdown linter, 50 rules, 38 auto-fixable
A fast Go Markdown linter and fixer with 50 rules (38 auto-fixable), shipped as a single zero-runtime-dependency binary. Distributed via Homebrew, Scoop, and GitHub Actions with SARIF output.
Constraint engine to curb AI hallucinations & prompt debt
A constraint engine for AI coding assistants that mitigates prompt debt and hallucinations with 100+ modular rules across security and code quality, delivered as context-aware guidance.
RSS reader that ranks a digest by how you actually read
A self-hostable RSS and newsletter reader with a behavioral, explainable ranking engine. Open source under AGPL-3.0.
Unified macOS system control with tiling & hot-reload
An open-source macOS tool providing Linux-level system control: a tiling engine, custom acceleration curves, and a background daemon for hot-reloading configurations across window management, input customization, and permission handling.
AI-powered explorer over 700k pgsql-hackers emails
An AI-powered reader for the pgsql-hackers mailing list archive, over 700,000 emails spanning decades. Semantic search, cited Q&A, and patch tracking by commitfest status.
Memtable, SSTables & Bloom filters in C, visualized
A from-scratch implementation of core LSM Tree components (Memtable, SSTables, and Bloom filters) in C, with a web visualization of operations and compaction.
Embeddable CPU/cache/scheduler simulations, 17 widgets
Canvas-based simulations for CPU pipelines, caches, schedulers, and storage. 17 widgets you can embed in Astro, React, Jekyll, or static HTML.
Docker env testing C code across 15 Linux distros
A Docker-based environment supporting 15 Linux distributions for testing C code compatibility, with an integrated VS Code extension for workflow management and automated testing.
AI extraction of educational clips from YouTube
An AI-powered tool for extracting educational video clips from YouTube using Groq's Whisper and LLaMA, with smart clip boundaries, dynamic currency support, and a microservice architecture for downloads.
Autonomous multi-agent system for end-to-end engineering
A multi-agent system built on LangGraph that plans, implements, audits, and verifies software end to end, instead of acting as autocomplete.
VS Code extension for structured AI prompts from code
A VS Code extension that generates structured AI prompts from code diagnostics and project context, with multi-provider switching and health checks across Groq and OpenAI.
Turns codebases into explorable knowledge graphs
A hierarchical code visualization system that transforms codebases into explorable knowledge graphs, integrating backend parsing with Tree-sitter and AST for real-time navigation.
AI-driven Doxygen-style comment generation
A CLI tool and Python library for AI-driven Doxygen-style comment generation, integrating Git change detection and PyPI packaging for cross-platform documentation automation.
Real-time search engine with Elasticsearch indexing
A real-time search engine for programming resources with WebSocket updates and intelligent crawling, built on a microservices architecture with Elasticsearch indexing.
HD video walkthroughs generated from GitHub repos
A web application that generates HD video walkthroughs from GitHub repositories, integrating AI narration, code highlighting, and dynamic execution visualizations.
Cross-platform C library for Kaggle datasets, 100% CI
A production-ready C library for programmatic Kaggle dataset downloads with a cross-platform CMake build system and 100% CI test coverage.
Detector & classifier deployed to Azure via Flask
A detector and classifier model that validates images and classifies tea-leaf quality, with the model served through a Flask API deployed to Microsoft Azure.
Tech stories, unfiltered
A publication for technical storytelling: engineering deep dives, post-mortems, and systems writing for developers.
Earlier open-source portfolio template, 11 GitHub stars
An older open-source portfolio template other developers have starred and forked. This site is the current portfolio.
Diamond Gate Protocol signal engine, correctness-first
A high-integrity signal-processing engine implementing the Diamond Gate Protocol, focused on "Correctness by Construction" with pure mathematical functions and validated schemas.
ML analysis predicting fatalities, MAE 53.09
An analysis of homelessness factors in Texas using probability testing and dimensionality reduction, predicting fatalities with a baseline MAE of 53.09.