Translational AI Scientist​/Engineer

Bio Age Labs (BIOA) is finding new ways to treat disease by targeting the mechanisms of aging, with the ultimate goal of increasing healthy human lifespan.

We are seeking a scientist/engineer with hands‑on expertise in building generative and agentic AI systems and a strong foundation in target discovery, drug discovery and translational science. You will design and deploy AI‑enabled systems that take a target and produce decision‑grade, evidence‑backed recommendations — from mechanism hypotheses to validation design — with every claim grounded in retrievable evidence and every gap stated explicitly, and that assess how likely a human‑derived signal is to hold up in the lab and beyond.

This role is ideal for someone who would rather build production AI systems for translational science than only run analyses, and who has enough hands‑on biology to know when a recommendation is scientifically sound and when it merely reads well. You will own the engineering of these systems end to end, from retrieval and orchestration to evaluation and deployment, and work as a peer with the scientists who act on their outputs.

• Design and deploy agentic AI systems that support scientific reasoning, hypothesis generation and evidence synthesis for the translational questions that follow target identification: mechanism and source of signal, indication selection, experimental design, reagent quality and translatability.
• Build systems that recommend how to test a target — experimental system (in vivo, ex vivo or in vitro), model, indication, endpoints tiered by translational relevance, intervention modality and study parameters — with each choice grounded in published precedent and with an explicit statement when no precedent exists.
• Develop tool‑using workflows that retrieve and integrate structured and unstructured evidence — knockout and perturbation phenotypes, endpoint precedent, published effect sizes, tool compound and reagent quality, prior programme outcomes, primary literature — with full provenance.
• Build the cross‑species layer: determine whether a target's human signal is reproducible in a model system, and identify the readouts that link experimental results back to the human cohort data.
• Assess translatability: develop evidence that a human‑derived signal will reproduce in vivo and onward, calibrated against targets with known preclinical and clinical outcomes, and feed that evidence back into target prioritisation.
• Build the evaluation framework for these systems — reference sets of targets with known experimental outcomes, metrics for citation quality and coverage, and calibration of translatability calls — with attention to scientific reliability and interpretability in decision‑critical settings.
• Encode domain rules on the meaning and reliability of each external source, so that absent, weak and contradicting evidence are handled distinctly and never collapsed.
• Partner with target biology and experimental teams so that recommended designs are usable by the people who run the studies, and incorporate their outcomes back into the system.
• Own these systems end to end (architecture, implementation, testing, deployment and monitoring) as maintainable software that scientists rely on day to day, not one‑off notebooks or prototypes.

• PhD with 2+ years of relevant experience,
  or Master's degree with 5+ years,
  or Bachelor's degree with 7+ years, in computer science, computational biology, biology, translation science or a related field. We welcome scientist‑first candidates with demonstrated ability to build production‑quality AI workflows, as well as engineer‑first candidates with deep translational‑biology judgment.

Preferred technical qualifications

• Hands‑on…