Director of AI Engineering

Recruits Lab connects specialized life-sciences and technology talent with organizations that need them—through relationships, not volume. You will turn that model into a durable advantage: production AI that finds passive experts, scores fit against culture and science, and shortens searches that used to take a quarter. This role decides whether we stay a high-touch boutique or become the data-driven standard in biotech, pharma, and technical staffing.

Performance Objectives

• Ship a production matching engine within 6 months that ranks passive life-sciences and tech candidates by verified skills, territory, and buyer-fit, lifting qualified shortlist conversion by 30%.
• Cut median time-to-calibrated-shortlist from 6–10 weeks to under 4 weeks on core searches by automating sourcing maps, named-account verification, and pipeline hygiene.
• Stand up an evaluation harness (precision, recall, bias, recruiter override rate) and run it on every model release so consultants trust the scores they show clients.
• Own the full stack from data contracts through inference: ingest CRM, outreach, and public signals; deploy retrieval + ranking; instrument latency, cost, and quality in production.
• Hire and lead a small AI/ML engineering team; set the roadmap so every sprint maps to a measurable recruiter or client outcome, not a demo.
• Partner with search consultants to encode “Tri-State / Midwest / Southeast” style territory and scientific fluency into features, not just keywords.
• Deliver a 12-month architecture that scales to new verticals (SaaS, oncology, digital pathology) without rewriting the core matching loop.

Environment & Resources

You report to the founding leadership of an 11–50 person, Fort Lauderdale–based firm. You work daily with recruiters who already win placements through passive, calibrated searches. Budget and tooling are sized for a focused production team—cloud, vector search, LLM APIs, and the CRM/ATS we live in. You set engineering standards; consultants remain the product experts.

Essential Qualifications

• Shipped ranking, retrieval, or matching systems used by non-engineers in production, with documented lifts in conversion or cycle time.
• Led 3–8 engineers building applied ML (embeddings, search, evaluation, LLM orchestration) rather than research-only work.
• Comfortable encoding domain rules (quota history, named accounts, scientific buyers) into features and guardrails.
• Proven ownership of data quality, evals, cost, and latency for models that affect revenue conversations.
• Can talk to life-sciences or technical commercial teams without a translator.

Role Selling Points

• Direct line from model to placement: your work sits in the searches that already close Area Sales Managers and similar specialist roles.
• Early-stage ownership: you define the AI stack instead of inheriting a bloated platform.
• Domain that compounds: life-sciences and tech talent maps are scarce, passive, and high-value—perfect for retrieval and ranking.
• Consultants who already think in outcomes, not headcount, so engineering is measured the same way.

If you want to build the matching layer that makes relationship-driven staffing scale, apply now. Show us a system you shipped that changed how experts get found.