8+ years in product · 10M+ users reached · $2M ARR · $2.8M annualized savings · 25+ features shipped
I ship AI products people can trust and use after launch.
At Trinka AI, I lead product strategy across AI writing tools, agent orchestration, and the model development lifecycle, with a cross-functional team of 20+.
I work as a product manager who stays deep in the AI system. I own problem definition, product direction, prioritization, and go-to-market. I stay close enough to architecture trade-offs, evaluation criteria, and failure modes to make sound product calls. When it helps the team move faster, I prototype and validate with AI-assisted development.
The metrics above cover products and initiatives I have helped build, lead, or scale. More context is in my portfolio and case studies.
- AI product strategy: product direction, roadmaps, prioritization, GTM, and cross-functional delivery
- Agentic product experiences: orchestration, tool use, human oversight, failure recovery, cost, and UX for complex agent behavior
- RAG and knowledge products: grounding, citations, document workflows, retrieval quality, and answer trust
- Evaluation and trust: quality gates, benchmarks, observability, safety, and production readiness
- Discovery to delivery: user research, PRDs, experiments, analytics, and shipping with eng, design, and research
Making AI products dependable after the demo:
- source-grounded generation users can verify
- agent evaluation and reliability as a product concern
- durable long-running workflows with clear human control points
- AI-native tools for product managers
- interfaces that make complex AI behavior understandable
My pinned repositories are experiments and products I use to pressure-test ideas in agentic AI, RAG, evaluation, and applied tooling.
Open to collaborating on AI product infrastructure, agent reliability, knowledge systems, and human-centred AI product design.