Lead AI Engineer

EPAM is looking for a Lead AI Engineer to serve as the technical owner of a connector implementation for Drug/Drug Interaction analysis. This role focuses on designing and building a Smart Connector using vanilla LangChain and LangGraph, architecting agentic orchestration, transforming natural-language questions into structured queries, validating a proof of concept, and ensuring statistical and clinical correctness in collaboration with Data Scientists and Pharma subject matter experts.

Design and build the Drug/Drug Interaction Smart Connector using vanilla LangChain and LangGraph.
Architect agentic orchestration that converts natural-language questions into structured queries.
Integrate structured queries with the PharmaPendium Advanced Search API.
Ensure retrieval through the FAERS module endpoint is reliable and traceable.
Lead a lightweight proof of concept to validate the implementation.
Resolve compatibility issues before they become blockers.
Build connector logic for candidate drug-pair generation ranked by interaction strength.
Implement MedDRA classification across generic and clinical phenotype terms.
Compute PRR/ROR disproportionality and produce traceable output tables.
Generate consistent outputs with case counts, statistical measures, and signal-only interpretation flags.
Write clean, well-documented code structured for Codex-assisted refactoring.
Collaborate with Data Scientists and Pharma SMEs to ensure statistical and clinical correctness.
Tune the connector against an agreed Golden Q&A set.
Support validation, acceptance baseline achievement, and handover.

5+ years of experience in AI engineering or related technical roles.
Expertise in vanilla LangChain and LangGraph.
Strong experience designing agentic orchestration for natural-language-to-structured-query workflows.
Experience integrating with APIs, preferably PharmaPendium Advanced Search API or similar search APIs.
Understanding of FAERS module endpoint integration or adverse event data retrieval is a strong advantage.
Knowledge of MedDRA classification across generic and clinical phenotype terms.
Ability to work with PRR/ROR disproportionality computation and statistical measures.
Experience leading proof-of-concept implementations.
Ability to resolve compatibility issues for future platform refactoring.
Strong clean-code and documentation practices.
Ability to collaborate with Data Scientists, Pharma SMEs, and engineering stakeholders.
English proficiency at B2 level or higher.

EPAM Systems — компания или платформа в сфере software engineering, IT и цифровой трансформации. Она предлагает разработку программного обеспечения, облачные решения, инфраструктуру, enterprise-системы, консалтинг или техническую поддержку для бизнеса, помогая клиентам, пользователям или партнёрам решать практические задачи с помощью современных технологий, экспертной поддержки и структурированных процессов. Компания была основана в 1993 году. Профиль компании на ANQ описывает её ключевые направления, продукты, услуги и роль на рынке.