1 day ago
Cambridge, MA, USAEntry Level
Responsibilities
- Design and develop agentic AI systems using large language models, multi-step reasoning, tool use, and MCP server integration.
- Build and optimize Retrieval-Augmented Generation pipelines for biomedical literature, clinical documents, and scientific publications.
- Deploy, monitor, and scale machine-learning models and AI agents using Google Cloud Platform, particularly Vertex AI.
- Implement automated and human-in-the-loop evaluation frameworks for LLMs and agentic workflows.
- Fine-tune or adapt foundation models for named entity recognition, classification, question answering, and summarization.
- Build end-to-end pipelines for data ingestion, chunking, embedding generation, vector indexing, retrieval, and search.
- Collaborate with bioinformatics teams and pharma scientists to translate scientific requirements into AI/ML solutions.
- Present technical results to cross-functional and non-technical stakeholders.
Requirements
- Bachelor’s or Master’s degree in Computer Science, Bioinformatics, Data Science, or a related quantitative field.
- 2–3 years of hands-on experience building AI/ML systems involving LLMs and NLP pipelines.
- Proficiency in Python and modern ML frameworks such as PyTorch, Hugging Face Transformers, or LangChain.
- Experience designing and deploying RAG architectures, including vector databases.
- Working knowledge of Google Cloud Platform services, especially Vertex AI, Cloud Run, and BigQuery.
- Prior experience in bioinformatics or working alongside pharma scientists on data-driven research is required.
- Exposure to bioinformatics workflows involving protein sciences is beneficial.
- Hands-on experience with agentic AI frameworks such as LangGraph, CrewAI, or AutoGen and LLM tool-use patterns.
- Experience fine-tuning or adapting large language models with LoRA, QLoRA, or PEFT is beneficial.
- Familiarity with MCP servers, multi-agent orchestration, and LLM evaluation or observability tools such as RAGAS, LangSmith, or Weights & Biases is beneficial.
- Must be based on the US East Coast for real-time collaboration.
- Ability to relocate to Upper Providence, Pennsylvania is an added advantage.
