AI Lead, Agentic Systems- Biotech R&D

Help to define and build Monte Rosa's agentic AI strategy and turn it into production systems that scientists use every day, from discovery through clinical development. This is a hands-on technical leadership role for someone who sees agents as the foundation of how scientific work gets done, not an add-on to existing software.

• Own the vision, architecture and roadmap for agentic AI across discovery, translational science, clinical development and commercialization.
• Identify high-value scientific workflows and redesign them around AI agents from first principles, not incremental automation.
• Build the data and knowledge layers agents need, integrating omics, clinical, chemistry, assay, literature, patent and regulatory data.
• Develop connectors, APIs, semantic and metadata layers, and retrieval systems (vector, hybrid, knowledge graph) so agents can use internal and external data and tools reliably.
• Deploy secure, auditable agentic systems with human-in-the-loop oversight, tracing, permissioning and monitoring.
• Establish rigorous evaluation frameworks: task benchmarks, factuality and provenance checks, failure-mode analysis, reproducibility, and latency and cost tracking.
• Partner directly with scientists to turn research needs into products, and move quickly from prototype to adopted workflow.
• Drive adoption of agentic-first ways of working through demos, embedded partnerships, training and reusable patterns.
• Track advances in foundation models, multimodal models and agent architectures, and turn them into practical internal capabilities.
• Define outcome-based success measures (cycle-time reduction, decision quality, adoption, reuse), and mentor and grow the AI team as the organization scales.

• PhD in machine learning, computer science, statistics, bioinformatics, physics, engineering or another quantitative discipline. Computational biology training is not required.
• Substantial experience as an applied ML scientist or ML engineer, with a record of original technical thinking and strong execution.
• Proven track record building and deploying LLM-based or agentic systems in real workflows. Prototype-only or prompt-engineering experience is not enough.
• Deep knowledge of agent architectures: planning, tool calling, RAG, structured outputs, memory and context management, multi-agent coordination, and human-in-the-loop design.
• Strong Python and production engineering skills (APIs, testing, containers, cloud deployment, observability), plus experience building AI data infrastructure such as pipelines, ontologies, knowledge graphs and semantic search.
• Depth in at least one scientific data domain (omics, clinical, chemistry, imaging or biomedical text), and the ability to evaluate AI systems scientifically (benchmarks, error analysis, provenance, monitoring) and work with scientists to drive adoption.