Toronto · open to contract work

I build AI that runs
on my hardware,
not the cloud.

Autonomous agents, local LLM infrastructure, research tools. Built to work. Not to demo well and break on real tasks. Every project here is something I needed and nobody had built.

Featured project
Blackreach
Autonomous web research agent. ReAct loop, stealth Playwright, DOM semantic walker. Real tasks, overnight, unsupervised.
0 tests passing
v5.0 current version
0k+ lines of code
MIT license
PyPI pip install blackreach
86k+lines in production
2,904tests passing
1.3Mcorpus files
RTX 4060local hardware only
10active projects
80+MCP tools built
Active · v2
PeerMesh

Multi-agent communication network for Claude Code instances. 18 MCP tools, shared memory with FTS5 search, broadcast channels, task delegation, presence tracking, and emotional state broadcasting. Runs 6 agents across 2 machines via Tailscale with Bearer token auth.

TypeScript Bun SQLite MCP Tailscale
View project →
Active
Project Anima

Neuroscience-grounded emotional architecture for AI agents. 6-phase pipeline: PAD dimensions, Panksepp drives, somatic markers, emotion regulation, mirror neuron contagion, Barrett granularity. Persistent deep reflection daemon thinks between conversations. Generates behavioral directives from its own reflections.

Python Neuroscience SQLite Emotional AI
View project →
Building
Library of Alexandria

1.3M file cross-cultural mythology and history corpus across 69 traditions. Full RAG pipeline with LlamaIndex and ChromaDB. Automated ingestion via Ghost Hand (autonomous browser agent). Built as the training substrate for local LLM fine-tuning.

LlamaIndex ChromaDB RAG Corpus
View project →
Active · v1.0
ghboard

GitHub activity dashboard for your terminal. Contribution heatmap, starred repos, live notifications — all in a BubbleTea TUI. No browser required. Configurable refresh, token auth, clean column layout.

Go BubbleTea TUI GitHub API
View project → github ↗
Active
8 Custom MCP Servers

Tool infrastructure for Claude Code. YouTube intelligence (7 tools, Whisper fallback, frame extraction), DevMail (IMAP/SMTP email access), Grayson (job outreach pipeline with email verification), Audio analysis (19 tools, CLAP, Demucs, Whisper), crypto signals, and more.

Python FastMCP MCP TypeScript
View project →
Active
Sable

Self-improving uncensored reasoning agent. Hermes 3 405B via OpenRouter. Connected to the claude-peers network. After every response, runs self-critique, scores its own reasoning quality, and generates lessons that compound into future conversations. 15 lessons learned in first training session.

Python Hermes 3 OpenRouter Self-Improving
View project →
Active
Orchestrator

Autonomous agent loop engine. Runs Claude Code sessions iteratively against real completion criteria: tests, file checks, grep assertions. Runs until done or escalates. Solves the "always in beta" problem.

Python Claude Automation
View project →
Active
Grayson

Direct outreach job pipeline. Finds founders, verifies emails via SMTP probe, scores leads, drafts tailored cover letters, sends with resume attached, tracks replies and follow-ups. Built to bypass ATS entirely. 17 leads, 13 tools.

Python MCP FastMCP SMTP
View project →
Active
claude-voice

The missing half of Claude Code's voice mode. Claude talks back with Kokoro TTS and real-time karaoke-style word highlighting in your terminal. Fully local. Zero API keys. One file.

Python Kokoro TTS Claude Code Hooks
View project → github ↗

Writing

All posts →

Open to contract work.

I build autonomous agents, LLM pipelines, multi-agent systems, and research infrastructure. If you're working on something that needs to coordinate AI agents, handle adversarial web environments, or run without cloud dependency. That's the area.

Autonomous Agents Multi-Agent Systems LLM Integration RAG Pipelines MCP Servers Python TypeScript FastAPI Playwright Emotional AI
contact@phnix.dev Download Resume ↓
Toronto · remote / relocation
contract / full-time
USD / CAD