I make LLM output in regulated financial workflows verifiable and measurable — citation-backed retrieval, self-verification loops, and an evaluation harness that measures claim-level grounding, unsupported-claim rate, and traceability.
I am the founder and CTO of MantaSol, a BFSI tech-compliance company focused on tech compliance and on building platforms for the safe use of advanced tech — AI agents and LLMs — in financial services. My work is advisory, documentation, and technical testing: it is not a legal opinion, and not a conformity assessment, certification, or attestation under any regulatory regime.
Alongside that, I build scalable data pipelines and ML systems that turn raw data into business value. With experience spanning data engineering, analytics, and machine learning, I focus on creating reliable, production-ready solutions, from ETL workflows and real-time streaming to model deployment and monitoring.