Pau, France • Founder @ Managed Code • Author of C# Interview Guide

I build AI‑native product systems on the .NET stack — knowledge platforms, GraphRAG pipelines, multi‑agent workflows, Orleans infrastructure and the developer tooling around them. My work blends product direction, software architecture and AI engineering into systems that carry context, take action and fit into real business workflows.

Quality without compromise. Production‑ready code from day one.

🧠 AIBase — enterprise knowledge platform for multi‑agent workflows, company knowledge, retrieval and operational execution.

⚙️ MCAF — framework for building software with coding agents through repository memory, explicit rules, docs, tests and repeatable delivery.

🌐 GraphRAG for .NET — .NET‑first GraphRAG stack with indexing workflows, semantic deduplication, community detection and pluggable graph stores.

🔗 SiloLinker — distributed link infrastructure for gated content, temporary access and monetisation flows.

📊 Keyload — monitoring, incident detection and operational visibility for websites, APIs and infrastructure.

These are examples of the kinds of systems I’m interested in building around AIBase:

• AI Patent Attorney — AI copilot for patent and trademark work, document handling, reporting and process clarity.
• Ticket Booking Copilot — multi‑step travel orchestration across flights, trains, schedules, budgets and live APIs.
• Food Delivery Copilot — compact orchestration system that validated AIBase under dynamic constraints and real‑time decisions.

• GraphRAG for .NET — graph‑based retrieval‑augmented generation for modern .NET workloads.
• MCAF — framework for building software products together with AI coding agents.
• MCPGateway — searchable MCP gateway for .NET.
• CodexSharpSDK — CLI‑first .NET SDK for OpenAI Codex workflows.
• ClaudeCodeSharpSDK — CLI‑first .NET SDK for Claude Code workflows.
• MarkItDown — C# tool for converting files and office documents to Markdown.

• Orleans.SignalR — scalable SignalR backplane powered by Orleans.
• Orleans.Identity — ASP.NET Identity integration for Orleans grains.
• Orleans.RateLimiting — rate limiting algorithms for distributed applications.
• Orleans.Balancer — load balancing and activation rebalancing across silos.
• Orleans.Indexing — search and indexing for grain state.

• Storage — universal interface for cloud blob storage providers.
• Communication — result pattern, pagination and explicit error handling for .NET.
• TimeSeries — high‑performance time‑bucket metrics primitives.

I write about AI workflows, product systems, software architecture, accessibility, performance and the operating reality of building systems that need to survive contact with production:

You can find all posts on my blog.

📖 C# Interview Guide — a practical guide for engineers who want sharper fundamentals, stronger interview thinking and a clearer view of how solid C# knowledge translates into real work.

Production‑ready code from the first commit.

Software should be engineered as if it will run in production tomorrow — because it often does. I care about clear architecture, explicit boundaries, meaningful names, strong typing, thorough tests and the discipline that keeps a system coherent when complexity pushes in from every side. Engineering shortcuts accumulate interest faster than any business debt, and fragile systems never recover their credibility.
If the code cannot handle real load, real change and real responsibility, it is not finished. Quality is not a nice‑to‑have — quality is the product.

• Strengthening AIBase as a memory‑and‑action layer for companies.
• Growing a .NET‑native stack around GraphRAG, coding agents and AI execution.
• Building systems that remain readable, testable and durable as scope expands.
• Writing and shipping in public with an uncompromising emphasis on quality.

Thanks for stopping by — let’s build things that last.