Head of Engineering JD
Company: Stealth Co.
Role: Head of Engineering (Full-Time)
Location: Flexible (US, preference for NYC)

Who We Are:

We’re building the AI-native operating system for litigation. Today, legal teams face millions of siloed documents across PDFs, emails, and spreadsheets—making facts hard to connect and knowledge fragile. Our platform turns this chaos into knowledge graphs rooted in structured ontologies, compounding insights with every case. The result: faster strategy, stronger arguments, and a lasting edge in high-stakes litigation.

What You’ll Do:

• Work with litigators and domain experts (scientists, engineers, other industry professionals) to understand workflows and design AI solutions for complex challenges
• Build and lead a high-performing product and engineering team
• Create AI agents that classify, extract, and connect facts from millions of documents into ontology design and knowledge graphs
• Communicate AI capabilities, tradeoffs, and technical reasoning to non-technical stakeholders in outcome-oriented terms
  

Who You Are:

• A builder-leader at the intersection of AI, product, and complex domains
• Experienced leading engineering in high-stakes environments (vertical SaaS, applied ML, Gen AI, or domain-heavy fields)
• Skilled at partnering with subject experts and translating pain points into technology
• Balances bold technical vision with clear customer impact
• Thrives in zero-to-one, fast-moving environments

Key Technical Responsibilities

• LLM Systems Architecture: Design and scale agentic systems using frameworks like LangGraph, PydanticAI, and Google ADK, leveraging modular reasoning pipelines (RAG, CoT, ReAct, planner–executor).
• Knowledge Infrastructure: Integrate vector databases (FAISS, Pinecone, pgvector), knowledge graphs, and full-text / enterprise search systems (e.g., Elastic, OpenSearch, Postgres FTS) to enable retrieval-augmented generation and persistent memory.
• Prompt & Evaluation Strategy: Develop robust prompt architectures, test harnesses, and evaluation loops to ensure reliability, consistency, and coverage.
• Engineering Architecture & Delivery: Define system architecture, technical standards, and development processes to support scalable, maintainable products.
• Team & Product Organization: Partner with product leadership to translate strategy into technical roadmaps, structure teams around clear ownership, and drive predictable, high-quality delivery.

Evaluation, Observability & Deployment:

• Build tools for trace inspection, state debugging, and hallucination detection
• Benchmark agent orchestration frameworks to assess performance, reliability, and ergonomics
• Deploy and monitor models in production with cloud-native tooling
• Build a modern AI stack with human-in-the-loop correction workflows, confidence scoring and well though QA frameworks.
  

Core Skills:

• 5-7+ years of Full-Stack experience
• 1+ year(s) experience in LLM agent development, prompt engineering, and use of frameworks like LangGraph, PydanticAI, and Google ADK
• Familiarity with vector databases (e.g., pgvector, Weaviate, ChromaDB)
• Production-level experience with FastAPI, Docker, and MLOps workflows
• Expertise in agentic IDEs (e.g., Windsurf, Cursor, Claude Code)
• Bachelor's or Master’s degree in Computer Science
  

Compensation:

• Base salary: $200,000–225,000
• Bonus based on performance
• Significant equity upside in platform