Available for collaboration

Building the Future of Agents

AI Agent Engineer · CLI Architect · Open Source

I design and build multi-agent systems, AI coding CLIs, and provider-agnostic LLM frameworks. Focused on making agents genuinely useful — real tools, real execution, real impact.

GitHub View Projects ↓

xclaw — AI programming CLI

$ xclaw --model "gpt-4o" --agent

# Multi-provider engine ready

✓ Anthropic · OpenAI · Gemini · Grok

# Agent loop active — tool execution

✓ Bash · Read · Edit · Write · Grep · Glob

# Permission system armed

✓ ask · auto-edit · full-auto

→ Analyzing codebase...

→ Reading src/engine/agent-loop.ts

→ Streaming tokens to TUI

// What I do

Core Competencies

End-to-end AI agent engineering: from provider abstraction layers to terminal UIs, from tool orchestration to multi-agent coordination.

🤖

Multi-Agent Systems

Collaborative agent frameworks with sub-agent spawning, task delegation, and inter-agent communication.

orchestration sub-agents MCP tool chains

⚡

CLI / TUI Engineering

Rich terminal interfaces with streaming output, markdown rendering, and real-time tool execution feedback.

TypeScript Ink / React Bun Vite

🧠

LLM Provider Integration

Provider-agnostic layers for Anthropic, OpenAI, Gemini, Grok — streaming, tool calling, function routing.

streaming tool-calling RAG alignment

// What I've built

Featured Projects

AI coding agents, multi-agent frameworks, developer tooling, and research tools.

⚡

xclaw

AI programming CLI — beyond the boundary of human and AI

Multi-provider AI coding agent with personality-driven TUI, real tool execution (Bash/Read/Edit/Write/Grep/Glob), agent loop with up to 20 rounds, streaming token output, permission system, and session persistence. Built with TypeScript + Bun + Vite.

TypeScriptBunInk TUIViteMulti-ProviderAgent Loop

🎹