I work with founders, CIOs, CTOs & boards focused on architecture, system design, and technical due diligence.
I approach agentic development through the discipline of Lean engineering: clear value streams, bounded roles, and continuous refactoring.
Agents need explicit job descriptions and hard guardrails. The workflow operates as a pull system: well-specified tasks wait on a board, where dedicated agents pull work within their scope, implement changes, and validate them against tests.
Without that structure, agents simply generate more code than anyone can review.
A founder needs senior technical leadership to design and deliver an MVP.
A production system starts to break under growth. Performance, security, scalability or infrastructure costs are becoming a problem.
Systems built by different organizations or with different technologies need to be integrated and work together.
A company needs to establish or evolve its technical organization, from architecture and engineering practices to pre-sales and professional services.
A board, founder or investor needs an independent assessment of technical readiness before making a significant investment or strategic decision.
A company has its software built by a supplier and needs someone on its side who can judge estimates, answers and releases.
Architecture sets the boundaries between components and the patterns that connect them, such as microservices or event-driven integration. Once the system is in production, it is the hardest part of the design to change.
System design maps the logical architecture onto a cloud or on-premises system, where it meets the constraints of physical reality. It decides how the system handles traffic peaks and server failures.
Software design is the structure of the code inside each component. I write it test-first and refactor it continuously, and every change goes through continuous integration. Code written by AI agents goes through the same tests before it is merged.
The previous system ran on PHP and Oracle, with no documentation, no automated tests and its database port open to the internet. The replacement runs as 18 services on AWS. It polls six traffic sources every 3 to 5 minutes and publishes about 10,000 events a day in three languages.
54% of database waits came from writing a log row for every API request: 2 billion rows, 13 TB. Caching token validation cut total SQL time by 77% and database CPU by 89% in load tests. I led the architecture review and the caching strategy with Redis on top of the database.
12 servers processed 6,000 files an hour, but searches took minutes and history was limited to two months. Our MongoDB proposal won a competitive PoC against SQL Server and Oracle. In the design benchmark, the new write path sustained 537 writes per second against the 300 required for ten times the volume.
The method lived in a workbook with its algorithms and past responses. I rewrote the scoring engine in Java and checked every result against the spreadsheet. The web application reached production less than three months after the first commit.
A single shared architecture for a luxury brand: frontend in Berlin, backend in Gurgaon, deployment in Shanghai, client in Geneva. Local compliance and localization were built into the design. It kept teams aligned across time zones.
I co-created RESTHeart, the open source backend for MongoDB, in 2014. RESTHeart Cloud, MobiStream and Sophia, a RAG platform, run on it.
Thirty minutes to describe the situation and work out whether I am the right person for it.
If you need an NDA before the call, send it with the invitation.