Manish Kumar

I’m a data scientist with 8 years of experience, including 6 years focused on US pharmaceutical commercial analytics using US claims and Rx data. I bring deep domain knowledge—TRx/NBRx dynamics, HCP-level analytics, line of therapy, patient journey, and marketing mix modeling—to make analytics decisions reliable and actionable.

In my current role as a Senior Data Scientist at Syneos Health, I build production AI agents for pharma analytics. I designed and developed an autonomous multi-agent analytics dashboard that turns natural-language questions into governed, self-serve outputs using a semantic model with entity resolution (ICD-10, NDC, CPT/HCPCS, and geography) and self-healing SQL generation.

I also architect multimodal RAG for payer persona research on primary payer content (PDF/Word/PPT plus interview transcripts), with metadata tagging, near-duplicate filtering, recency-aware conflict handling, and persona-grounded Q&A with per-answer source citations. Earlier, I delivered marketing mix modeling at Axtria Ingenious Insights, and I built a COPE prediction tool at Xceedance Consulting that improved loss-prediction accuracy through FEMA flood-zone mapping.