I lead product, engineering, and AI platform strategy at the point where ambiguity becomes execution.
Over 25 years in software, I have led and advised product, engineering, and strategy teams through platform bets, architecture risk, organizational complexity, and delivery pressure. My work is not only about building systems; it is about deciding which systems are worth building, aligning people around them, and carrying them into production.
I think like a product owner, operate like a strategist, and stay technical enough to know where the system will break. I care about teams and agents that can take responsibility in real codebases, not just look impressive in a terminal.
In 2026, while continuing my professional work, I made a deliberate turn back toward open source: to help public engineering teams, contribute to systems I respect, and make my judgment visible in the open. Occasionally loud in Turkish.
- 🪑 claude-roundtable — Multi-agent governance for Claude Code. Council deliberates, votes, dispatches, and enforces quality gates.
- ⏱️ leyla-scheduler — Durable, session-aware task scheduler in Rust. Jobs survive disconnects.
- 🔎 skill-hunter — Pre-execution layer that makes agents check for existing skills before building from scratch.
- 🛡️ promptguard — Audits prompts as behavioral contracts. Pre-write guard for agents that ship code.
- 🧠 recall-mcp — Shared brain for AI agents. SQLite-backed persistent memory over MCP.
- 👵 moooom-claude — 10 cultures of moms nagging your Claude Code to drink water and sit up straight.
- Leading or advising product, engineering, and strategy teams larger than 15 people for nearly 15 years.
- Leading product and engineering teams through ambiguity, delivery pressure, and architecture risk.
- Advising multi-disciplinary teams where product direction, technical strategy, and execution discipline have to move together.
- Turning fuzzy product direction into executable systems.
- Seeing the strategy, user workflow, and failure mode in the same frame.
- Choosing the next useful move when everyone else is still debating abstractions.
- Raising the bar without slowing the room down.
- Turning vague bug reports into concrete failing cases.
- Finding the smallest patch that a maintainer can actually review.
- Building guardrails so agents do not silently improvise their way into production.
- Staying with the boring part: CI, review comments, rework, and merge readiness.
I come from a mathematical way of thinking: definitions first, invariants second, implementation third. My research line connects product strategy with formal systems for AI agents:
- TeserracT — policy-constrained stochastic control, type-level governance, and market economics for multi-tenant agent platforms.
- Miray Tesseractic Mathematics — a semantic mathematics program for LLM meaning across geometry, energy, topology, type theory, memory, and human impact.
- Gated Persona Intelligence — stability-controlled persona memory for LLM systems; entropy gates, sigmoid smoothing, hysteresis, and deterministic cognitive state engines that reduce drift in long-running agent interactions.
That background shapes how I design products: formal enough to reason about, practical enough to ship.
The interesting failures in agent systems are not capability gaps — they are accountability gaps. Agents do not push back on scope. Memories drift into summaries. Self-correction is theatre without external validators. I write infrastructure that turns those soft failures into hard ones: a council that has to vote, a scheduler that has to resume, a guard that has to read the contract before the agent ships.
The goal is simple and hard: agents that can survive contact with real repositories, real tests, and real maintainers.
Product leadership · Technical strategy · Agent systems · Rust · Go · TypeScript / Next.js · Python · .NET / C# · SQLite · MCP · LLM evaluation
"Code is universal. Some commit messages are in Turkish. Make peace with it."