AI engineer and cloud architect with 16+ years building and scaling production software systems. I help startups and product teams take LLM features from prototype to production: combining Azure OpenAI, Java, Python, TypeScript, and cloud-native architecture to ship systems that stay reliable under real traffic.
I specialise in the engineering layer around LLMs, where most AI projects actually fail: inference isolation so model latency and outages don't take down core services, structured prompt engineering with response validation and guardrails, low-latency service integration, and cost control under variable AI workloads. Alongside that, the fundamentals: LLM-powered APIs, backend services, and event-driven microservices on Docker, Kubernetes, Azure, Kafka, and CI/CD.
Currently at Microsoft, bringing enterprise-grade engineering discipline to fast-moving startup environments. I work comfortably in small distributed teams, communicate clearly with technical and non-technical stakeholders, and focus on AI features that reach users rather than demos.