AI Systems
Production LLM inference, agentic interfaces, tool integration, evaluation, and safe operational boundaries.
Exploring Staff AI Engineer opportunities 2026
Building useful AI systems
for the real world.
I’m an engineer who enjoys turning complex ideas into dependable products across AI, distributed systems, and fintech.
10+Years engineering
6Engineering roles
10+Products & platforms
01 · Capabilities
What I’ve learned
through building.
Distributed Platforms
Stateful orchestration, resilient execution, and on-chain integrations designed for throughput, isolation, and observability.
Optimization
Search and simulation systems that turn large strategy spaces into explainable, testable decisions.
Product Delivery
I work from early discovery and system design through hands-on implementation, launch, and ongoing improvement.
02 · Selected work
Systems I’ve built
and worked on.
Projects where I contributed across architecture, implementation, and product decisions—from regulated payments to trading infrastructure.
Managing
Staff AI Engineer–oriented trading platform: agents (Semantic Kernel, MCP, Hermes, n8n), Optuna research pipelines, Spotlight/news/regime context, and distributed live execution.
Xpollens
Production payment platform modernization—SEPA Instant, virtual IBANs, partner tooling, and operational monitoring that built reliability habits for Staff-level systems work.
Enterprise & Product Engineering
Financial analytics, ERP, derivatives platforms, and independent product delivery—foundation in regulated domains, data products, and full ownership of outcomes.
03 · Field notes
Notes from
the work.
Ideas and lessons from designing AI products and distributed systems.
01
Designing AI-Assisted Quantitative Systems
Where AI belongs—and where it does not—in strategy design, validation, and execution.
02Distributed Execution with Microsoft Orleans
A practical architecture for isolating stateful workloads while scaling execution safely.
03Production LLM Integration Beyond the Demo
Engineering boundaries, evaluation, observability, and failure handling for useful LLM features.