We are partnering with a company that is the scientific intelligence layer powering life sciences dealmaking. We combine expert-developed diligence frameworks, AI reasoning trained on drug development, and a secure, unified evidence base to help pharma, biotech, and deal teams move assets from evaluation to term sheet — faster and with more confidence.
The Role
We are looking for an AI Engineer to build the reasoning systems at the core of our platform: AI agents that understand drug development risk, therapeutic-area- and modality-specific models, and end-to-end pipelines that turn fragmented scientific evidence into decision-ready outputs. You'll work at the intersection of applied AI and life sciences, building systems that require both technical rigor and grounding in real scientific evidence — with human-in-the-loop validation and full data provenance baked in from day one.
***This is a hybrid role with 2 days a week onsite in Manhattan***
What You'll Do
- Design, build, and ship AI agents and reasoning pipelines that evaluate biological mechanism validity, translational/regulatory/clinical risk, and deal-relevant scientific evidence
- Develop and fine-tune models trained across therapeutic areas, modalities, and the drug development lifecycle
- Build retrieval and evidence-grounding systems that connect proprietary, public, and internal data sources into a single, auditable evidence base
- Implement human-in-the-loop validation workflows and maintain end-to-end data provenance across every model output
- Partner closely with scientific and product teams to translate expert diligence frameworks (PTRS, NPV, development scenario modeling) into production AI systems
- Own performance, reliability, and evaluation of models and agents in production, iterating quickly based on real deal usage
- Work within a security-first architecture (single-tenant, SSO, encryption at rest/in transit, ISO/IEC 42001:2023-aligned) — your data and model choices will need to hold up to pharma-grade scrutiny
What We're Looking For
- Strong software engineering fundamentals, with production experience building and deploying AI/ML systems (not just notebooks/prototypes)
- Hands-on experience with LLM-based systems — agent architectures, RAG/retrieval pipelines, evaluation frameworks, prompt/model tuning
- Comfort working with ambiguous, evidence-heavy, semi-structured data and translating expert domain frameworks into reliable systems
- A bias toward shipping — comfortable in a small, fast-moving team where you own problems end-to-end
- Genuine interest in life sciences, drug development, or healthcare AI (prior experience in the space is a plus, not a requirement)
- Excellent judgment around data provenance, auditability, and security — this product handles some of the most sensitive data in the industry
Nice to Have
- Experience with biomedical/scientific NLP, clinical trial data, or drug development datasets
- Background in a regulated or high-trust data environment (healthcare, fintech, etc.)
- Experience building agentic systems with tool use / multi-step reasoning