At Future Secure AI, I lead technical delivery of the internal AI agents platform and observability tooling while staying hands-on. I coordinate developers and QA, guide architecture and code reviews, and work directly with customer stakeholders on live workflows and integration issues.
I built voice agents for internal hospital communications and educational agents that help students revisit classroom material using teacher-maintained lesson context. I develop LLM features used by more than 10k active users, including RAG and multi-provider integrations.
I evaluated agent responses against a golden dataset and implemented input guardrails to mitigate prompt injection through voice. I also built an internal MCP server for metric queries, log and trace correlation, and alert routing; fixed a critical race condition and improved HTTP endpoint latency by up to 40%.
Previously, at AIZZY, I built APIs and React interfaces for an AI assistants platform used by more than 10k active users, along with document RAG pipelines and asynchronous processing. At Maker Experts, I built React Native applications for public-sector agencies and owned App Store and Play Store releases.
