I build developer tools around Artificial Intelligence, LLMs and agentic workflows — from terminal coding agents and MCP integrations to local AI systems and model orchestration.
I'm a developer focused on the intersection of software engineering and Artificial Intelligence.
Most of my work revolves around building practical infrastructure for AI-assisted development: coding agents, terminal interfaces, MCP servers, multi-model orchestration, local LLM tooling and developer experience.
I'm also currently studying Artificial Intelligence Theory to strengthen the foundations behind the systems I build — including machine learning, neural networks, search algorithms, optimization, model evaluation and reinforcement learning.
I don't want to only integrate AI APIs. I want to understand how intelligent systems work, how they fail, and how to build better tools around them.
Open-source AI coding assistant that lives in your terminal.
A full terminal-based coding environment with agentic workflows and a rich TUI, designed around modern LLM-powered software development.
- Agentic coding workflows
- Fullscreen terminal UI
- Multiple AI providers
- MCP support
- Sub-agents
- Optional LSP navigation
- Memory and sessions
- TypeScript + Bun
A local MCP orchestration layer for AI coding agents.
Allows coding agents to consult external models, compare answers, run adversarial checks and trigger scoped code or security reviews without leaving the development workflow.
- Multi-model consultation
- Parallel answer comparison
- Adversarial verification
- Code review orchestration
- Security-oriented review flows
- Structured MCP tooling
An autonomous local-LLM agent with tools, planning and optional RAG.
Lyre gives local models running through LM Studio, Ollama or OpenAI-compatible servers the ability to work with files and execute real tasks inside a controlled environment.
- File-system tools
- Multi-step planning
- Undo snapshots
- Context management
- Optional RAG with semantic search
- Local-model support
Artificial Intelligence Theory
├── Machine Learning
├── Neural Networks
│ ├── Perceptrons
│ ├── Activation Functions
│ └── Backpropagation
├── Search Algorithms
│ ├── BFS
│ └── DFS
├── Optimization
├── Model Evaluation
│ ├── Confusion Matrices
│ ├── Precision / Recall
│ └── Overfitting / Regularization
├── Reinforcement Learning
└── Foundations of modern AI systems
My goal is to connect theory with engineering: understanding the principles behind AI systems while building practical tools that use them.
|
|
Languages
TypeScript · JavaScript · Python
Runtime & Backend
Node.js · Bun · FastAPI
AI & Agent tooling
LLMs · AI Agents · MCP · RAG · Coding Agents
Development
Git · GitHub · Linux · Docker
- Open AI Gateway — API gateway for AI development workflows
- Codex Gateway — experiments around model and coding-agent gateways
- Nebraska — additional open-source experimentation
