AI Engineer

Recruits Lab exists to place the right life-sciences and technology talent faster than anyone else—by treating recruiting as a data problem, not a volume problem. This role builds the systems that turn passive-talent maps, named-account verification, and calibrated shortlists into a repeatable, defensible advantage for every search we run.

• Ship production matching and ranking models that raise first-interview-to-offer conversion by 20% within two quarters.
• Automate territory and buyer-relationship mapping so recruiters start every search with a verified passive list instead of a job-board scrape.
• Deliver an evaluation pipeline that scores quota history, deal size, and back-channel signal in hours, not days.
• Cut time-to-calibrated-shortlist from weeks to days on life-sciences commercial and technical searches.
• Own model monitoring, bias checks, and data quality so every recommendation we send a client is auditable.
• Partner with search leads to instrument every placement so the next model learns from what actually closed.

You report to the founding team in a 11–50 person, early-stage firm. You will work directly with recruiters who live in the life-sciences and SaaS talent maps, using whatever stack gets the job done (Python, modern LLM/RAG tooling, internal CRM and ATS data). Budget and cloud resources follow results, not process theater.

• Shipped ML or LLM systems that measurably changed a business process, not just a demo.
• Comfort extracting signal from messy, sparse, relationship-heavy data (CRM notes, call outcomes, named-account lists).
• Ability to sit with a recruiter, hear the real failure mode, and ship a fix they will actually use.
• Strong Python plus practical experience with retrieval, ranking, or evaluation pipelines.
• Bias toward production, monitoring, and iteration over research papers.

• Direct line of sight from model to placement—no layers, no vanity metrics.
• Domain that is still analog: life-sciences field sales and biotech talent is won at the sourcing stage, not the application stage.
• Early-stage ownership: you define the AI stack for a firm whose entire brand is data-driven recruiting.
• Work that compounds—every closed search makes the next one faster and more accurate.

If you want to build the intelligence layer that actually changes how specialized recruiting gets done, send a note that shows you already think in outcomes. We read every one.