Senior Applied Scientist, Generative AI
Liberty Mutual19 hours ago
Base Salary
$120k - $225k/yr
Responsibilities
- Design, fine-tune, evaluate, and deploy language-model solutions from research and experimentation through production implementation.
- Fine-tune and distill small language models for domain-specific insurance tasks while balancing accuracy, latency, and cost.
- Build GenAI pipelines including retrieval-augmented generation, corrective RAG, and agentic orchestration with tool use, planning loops, and memory.
- Develop scalable data, document, and embedding pipelines using MLOps and LLMOps practices for reproducibility, deployment, and monitoring.
- Design evaluation suites and guardrails covering groundedness, accuracy, safety, and regression testing.
- Containerize and ship solutions on Kubernetes and partner with engineering teams to operationalize them in production.
- Research and prototype methods for training, adapting, and evaluating generative models and share findings with the team.
- Present technical findings and recommendations to technical and non-technical stakeholders and participate in cross-functional working groups.
Requirements
- Ph.D. in Statistics, Computer Science, Mathematics, Economics, Actuarial Science, or a related quantitative field with 2+ years of relevant experience; or a Master's degree with 4+ years; or a Bachelor's degree with 6+ years.
- Expertise with transformer-based language models, including LoRA, QLoRA, and instruction tuning.
- Hands-on experience with retrieval design, chunking, embedding strategies, vector search, and agentic tool use.
- Strong foundations in machine learning, statistics, experimental design, model evaluation, and generative-output assessment.
- Proficiency in Python, MLOps practices, Git, collaborative development workflows, and model versioning or experiment tracking.
- Proficiency with PyTorch and GenAI tools such as Hugging Face, LangChain, LlamaIndex, or LangGraph.
- Experience with workflow orchestration tools such as Airflow or Luigi.
- Experience with Docker, CI/CD pipelines, Kubernetes, Helm, and GPU-backed services.
- Understanding of GPU acceleration, distributed training, and inference optimization including mixed precision, quantization, KV caching, and request batching.
- Experience with multimodal models, cross-modal retrieval, and vision-language models is preferred.
- Experience with insurance data or regulated-industry requirements such as responsible AI review, data governance, and auditability is preferred.
- Ability to communicate complex technical concepts, advance research from ideation to implementation, build relationships, and present recommendations to diverse audiences.
Benefits
- Hybrid schedule of two office days per week for candidates within 50 miles of Boston, Portsmouth, Seattle, Columbus, or Plano; otherwise remote with occasional travel.
- Comprehensive benefits, workplace flexibility, professional development opportunities, and Employee Resource Groups.
- Inclusive workplace culture with support for employee well-being and meaningful business impact.
Tech Stack
Categories
About Liberty Mutual
Liberty Mutual is a global mutual insurance company that provides property-casualty coverage and related services to individuals and businesses, including auto, home, renters, small commercial, specialty, surety, and reinsurance. It distributes policies directly and through brokers and partners, and also manages an investment arm deploying long-term capital. Founded in 1912 and headquartered in Boston, it is a Fortune 100 insurer operating in more than 25 countries.