Senior Scientist - GenAI Evaluation

At Johnson & Johnson, we believe health is everything. Our strength in healthcare innovation empowers us to build a world where complex diseases are prevented, treated, and cured, where treatments are smarter and less invasive, and solutions are personal. Through our expertise in Innovative Medicine and MedTech, we are uniquely positioned to innovate across the full spectrum of healthcare solutions today to deliver the breakthroughs of tomorrow, and profoundly impact health for humanity. Learn more at jnj.com.

As guided by Our Credo, Johnson & Johnson is responsible to our employees who work with us throughout the world. We provide an inclusive work environment where each person is considered as an individual. At Johnson & Johnson, we respect the diversity and dignity of our employees and recognize their merit.

Job Function:

Job Sub Function:

Job Category:

All Job Posting Locations:

Job Description:

At J&J we are building Generative AI solutions to support pharmaceutical R&D — literature review, evidence synthesis, document Q&A, therapeutic area knowledge search, translational science workflows, and R&D decision support. These systems need to be evaluated before teams rely on them in scientific workflows. In pharma, a useful AI response depends on the question, user, source material, therapeutic area, and risk of error — so quality must be measurable, repeatable, traceable, and scientifically defensible.

As a Senior Scientist in our Generative AI Evaluation & Standards (EQS) function within Data Science and Digital Health, you will design, build, and run the evaluation methods we depend on to assess these systems across R&D. You will author rubrics, curate benchmark and golden datasets, validate AI judges, and produce the quality readouts that inform release decisions. Want to shape how a top pharmaceutical company determines whether its AI is ready for real scientific work? This is that role!

KEY RESPONSIBILITIES:

We need someone who can build the evaluation assets our teams count on, and continuously improve them based on what we learn. Day to day, you will:

• Design, build, and maintain automated evaluation pipelines for LLM quality, RAG performance, agent reliability, safety, and scientific accuracy.
• Author evaluation rubrics and scoring criteria, curate golden and synthetic datasets with domain experts, and maintain our registry of reusable evaluation assets.
• Validate AI judges against human expert agreement and run model, prompt, retriever, and agent benchmarks that produce standardized quality readouts.
• Analyze failure patterns — hallucination, unsupported claims, weak traceability — and turn findings into actionable recommendations.
• Develop therapeutic-area-specific evaluation criteria with scientific, clinical, and regulatory partners, refining them based on real-world feedback.
• Design evaluation methods for scientific reasoning, evidence synthesis, and hypothesis quality — where generic benchmarks fall short.
• Build evaluation tooling and reusable patterns that enable other teams to self-serve.

QUALIFICATIONS

Education:

Master's degree in AI/ML, Computer Science, Data Science, Computational Biology, Bioinformatics, Biomedical Engineering, Applied Mathematics, Biostatistics, or a related field required. PhD preferred.

EXPERIENCE AND SKILLS:

Required:

We are looking for someone with 6+ years of hands-on experience in AI/ML evaluation or data science (Master's) or 3+ years of industry experience (PhD). You should have experience designing and running evaluation frameworks, scientific benchmarks, or quality assessments for AI/ML systems, and hands-on work with generative AI — large language models, retrieval-augmented generation, agentic frameworks, and prompt engineering. We also value strong proficiency in Python and modern AI/ML tooling (evaluation harnesses, embedding models, vector databases, LLM APIs), the ability to translate expert scientific judgment into measurable criteria, rubrics, and reproducible protocols, and a collaborative, self-driven approach to working across multidisciplinary teams.

PREFERRED:

We would love to find someone who also brings experience with AI/ML evaluation in regulated environments (FDA, EMA, or equivalent), understanding of the drug development pipeline and biomedical data types, or domain expertise in oncology, immunology, or neuroscience. Experience designing or validating LLM-as-judge systems, implementing CI/CD evaluation pipelines, or publications in AI evaluation, NLP, or biomedical informatics are all a plus.

OTHER:

English proficiency is required (written and verbal). This is a hybrid role based in Madrid or Barcelona, with limited travel (<10%), primarily within Europe.

Required Skills:

Preferred Skills:

The anticipated base pay range for this position is:

Benefits:

In addition to base pay, we offer the following benefits*: an annual bonus with set target (% of pay) depending on pay grade / location, where the actual amount is based on the employees’ and companies’ performance of the previous calendar year, or sales commissions. Moreover, we offer vacation days, parental leave for a minimum of 12 weeks, bereavement leave, caregiver leave, volunteer leave, well-being reimbursement, programs for financial, physical and mental health. We also offer service anniversary and recognition awards, and subject to the terms of their respective plans, employees - and in some location’s eligible dependents - can participate in several insurance plans. For more information, visit Employee benefits | Supporting well-being & career growth | Johnson & Johnson Careers.

*This is for informative purposes only. Amounts and actual benefits may vary by location and are subject to change.