AI Engineer

This is a fully remote position, open to applicants in India.

📋 Description

• Develop hybrid retrieval systems utilizing BM25, sparse, and dense embeddings, with reranking capabilities over millions of papers from sources such as PubMed, OpenAlex, and PMC full text.

• Enhance the matching process among papers, protocol steps, JoVE videos, and timestamps, taking into account organism, technique, and study type awareness.

• Create RAG and agentic pipelines for cited synthesis, Methods Q&A, and troubleshooting, ensuring every claim is traceable to a source.

• Extract method steps, reagents, and parameters from papers, linking each step to the corresponding video chapter.

• Develop evaluation sets in collaboration with scientific editors, tracking relevance, citation accuracy, hallucination rates, and match quality to manage releases.

• Deploy and manage models and LLM services in production environments while monitoring latency and cost per query.

• Select between hosted and open models based on project needs.

• Assist in creating APIs and an MCP server for AI assistants, ELNs, and publisher widgets to leverage JoVE’s methods data.

• Analyze the search, video-click, and watch funnel in partnership with product teams, aiming to enhance ranking.

⛳️ Requirements

• Minimum of 4 years of experience in software or ML engineering, with at least 2 years dedicated to deploying LLM, search, or NLP features into production.

• Proficient in Python with strong software engineering practices, including testing, code review, and CI/CD.

• Practical experience in building RAG systems, including chunking, embeddings, hybrid retrieval, reranking, and prompt design.

• Familiarity with vector or search engines such as Elasticsearch/OpenSearch, Milvus, pgvector, Qdrant, or Pinecone.

• Experience working with LLM APIs (Anthropic, OpenAI, or similar) and open models (Hugging Face, vLLM).

• Proven ability to create offline evaluations and utilize metrics, rather than demos, to inform shipping decisions.

• Experience with agent frameworks, tool utilization, or developing MCP servers.

• Comfortable working with cloud infrastructure (AWS), Docker, and API design.

• Effective communication skills with non-engineers, including scientists and editors.

• Background in scholarly data, including PubMed, OpenAlex, Semantic Scholar, PMC, or entity extraction related to genes, organisms, and methods.

• Experience in multimodal projects involving video transcripts, timestamp alignment, or image and video embeddings.

• Skills in fine-tuning embedding or reranker models, or learning-to-rank methodologies.

• Experience managing large-scale indexing (100M+ documents) and vector quantization techniques.

🏝️ Benefits

• A competitive compensation package.

• Fully remote work environment!

• Opportunity to make a direct impact in accelerating scientific research and education.

• Collaborate with global teams in a setting that encourages innovation and teamwork.

• Our strong culture of promoting from within provides a clear pathway for career advancement.