AI engineering student building reliable autonomous agents and developer tools.
Creator of Grinta,
a local-first coding agent that completed a 4 h 33 min autonomous run across
16,393 recorded events — without additional prompts.
Grinta · Demo · LinkedIn · Email
Provider-agnostic terminal agent for long-running repository work. Plans, edits, executes commands, observes failures, and validates outcomes autonomously. Persists an event ledger with checkpoints, replay, and durable recovery. Integrates local LSP and DAP tooling for code navigation and debugging.
v1.0.0 · 4h33 autonomous run · 16,393 recorded events · Raft case study: 39/39 tests · PyPI / Docker
21 runnable Python implementations covering planning, reflection, tool use, RAG, memory, evaluation, multi-agent coordination, MCP, and human-in-the-loop.
32 ★ · 9 forks
Full-stack multi-agent planning workspace using Next.js, TypeScript, Zod, SSE, Prisma, and Gemini. Coordinates specialized AI roles across product planning, architecture, security, implementation, and QA — outputs validated as structured artifacts.
AI-powered compliance validation for financial documents. Analyzes PPTX / PDF / DOCX for ESG / SFDR compliance, data consistency, chart accuracy, and disclaimer verification. Python backend with Next.js dashboard and REST API.
Blazing-fast trigram-based code indexer for sub-millisecond substring search in large codebases. Built in Rust for LLM agents and developer tools.
Stack: Python · TypeScript · Rust · Next.js · FastAPI · LangChain · Asyncio · Docker
Agentic AI Intern at Sofrecom Tunisia · AI Engineering at ESPRIT · Graduating 2027
Open to internship and collaboration opportunities in AI engineering and developer tools.



