Ai Lead Engineer

About Playbook :

At Playbook, we're building the control layer for agentic automation to govern the workforce of the future. We enable companies to inject rule-based behavior into autonomous AI agents by translating regulatory and operational directives—drawn from SOPs, policies, and best practices—into structured guidance. These rules are encoded into retrieval-augmented knowledge graphs that seamlessly integrate into any language model.The problem :

Companies today spend up to 25% of their revenue on compliance activities like training, audits, and documentation, yet failures are still common. We aim to eliminate that overhead by replacing the compliance function with a real-time control layer that governs AI agents on the ground. Our initial focus is on healthcare, especially the pharmaceutical industry, where legacy SOPs govern critical processes vital for quality, traceability, and patient safety.Founded in 2022,

Playbook is a post-revenue company serving over 1000 quality leaders in pharma and MedTech. We are a team of 22 (including 12 engineers), backed by leading VCs like Capnamic and 14 Peaks Capital, with recent seed funding in Q1 2025.About the Role :

AI Solution Architect :

As our AI Lead Engineer, you will shape the design and delivery of the core intelligence layer behind Playbook's platform. Your work will include encoding complex regulatory rules into structured knowledge graphs and embedding them into next-gen LLM-based agents. You will architect and engineer solutions that are foundational for scaling our technology across industries, working from system design to hands-on development—bridging AI architecture, product thinking, and enterprise integration. This role is ideal for someone eager to take ownership of technical direction, shape infrastructure, and make impactful decisions affecting clients, models, and products.What You’ll Do :

Strategic Planning and Roadmap DevelopmentArchitect and build cloud-based infrastructure on AWSDevelop RAG pipelines translating regulatory content into knowledge graphs and control logicDesign real-time orchestration patterns for agent frameworks like LangGraph, LangChain, CrewAI, AutogenCreate scalable, modular components using vector databases, graph databases, semantic search, and transformersOwn decision-making around system architecture, data pipelines, model integration, and retrieval strategiesCollaborate with clients and internal teams to translate requirements into AI-driven compliance layersPrototype, validate, and iterate quickly in an agile environmentInfluence and evolve the platform’s technical strategy with product and engineering teamsQualifications & Requirements :

Minimum 5 years of experience in AI / ML systems in production, preferably in regulated settingsAt least 5 years of experience with AWS cloud infrastructure (Sagemaker, DynamoDB, etc.)Strong expertise with LLMs, RAG pipelines, and knowledge graph designProficiency in Python, JavaScript, LangGraph, LangChain, PyTorch, TensorFlow, OpenAI APIs, Hugging FaceDeep understanding of retrieval systems, semantic embeddings, vector databases (FAISS, Weaviate, Pinecone), and orchestration frameworksExperience translating complex rules into formalized logicDesigning scalable containerized environments (Docker / Kubernetes)Excellent communication skills for cross-functional collaborationBonus Points :

Experience with document parsing, OCR, NLP pipelinesBackground in knowledge representation, ontology engineering, or symbolic AIOpen-source contributions or published researchExposure to agent orchestration frameworks like CrewAI, AutogenPerks & Compensation :

Competitive salary with equity stakeInfluence in a deep tech company with enterprise traction and growth potentialFlexible work setup—remote / hybrid optionsWork with a high-caliber team of engineers and researchersOpportunities for leadership and industry impactAccess to conferences, learning budgets, and essential toolsThe Stack You’ll Touch :

LangGraph, LangChain, CrewAI, Autogen, OpenAI APIs, LlamaIndexPython, JavaScript, FastAPI, Neo4j, FAISS, WeaviateAWS, Docker, Postgres, GitHub Copilot, Notion

J-18808-Ljbffr

#J-18808-Ljbffr