What are the responsibilities and job description for the Founding AI/ML Engineer position at Stealth?
We are a pre-seed funded reg tech startup ($1.5M raised), building Carta for regulated companies. We’re creating AI agents that draft, verify, and benchmark public companies and registered fund's filings with human-level reasoning and legal defensibility.
- Multi-agent orchestration (serial parallel) using LangGraph, MCP, and PydanticAI
- Graph-based retrieval (Graph RAG) across filings, entities, and historical disclosures
- Reinforcement learning for consistency, hallucination reduction, and reward-based accuracy
- Deep integration with corporate data systems (ERP, 365, equity management) to create a single source of truth
The Role
We’re hiring a Founding AI/ML Engineer to own architecture, algorithms, and reliability.
You’ll design multi-agent systems that reason over structured unstructured data and deliver auditable, regulator-ready outputs.
This role sits at the intersection of applied research, product, and infrastructure — you’ll ship code that defines the next category of enterprise compliance software.
What You’ll Do
- Architect multi-agent systems for reasoning and verification in high-stakes regulatory workflows
- Build graph-based retrieval and long-context scaffolding for fact consistency and cross-document reasoning
- Develop evaluation and reinforcement learning frameworks to optimize accuracy and minimize hallucination
- Create domain-specific prompting and chain-of-thought patterns for corporate filings and risk disclosures
- Translate legal and financial reasoning into programmable state machines and decision DAGs
- Ship fast — from prototype to production — with full ownership over latency, reliability, and evaluation
What We’re Looking For
- 3 years in ML/AI, with strong grounding in transformer architectures, RLHF, or retrieval-augmented systems
- Experience building or scaling multi-agent architectures or graph-based reasoning systems
- Proficiency in Python (LLM frameworks, orchestration tools, cloud pipelines)
- Experience working in high-security or high-accuracy domains (finance, legal, healthcare)
- Bias toward speed, clarity, and ownership — you think like a founder, not an employee
Nice to Have
- Experience with LangGraph, PydanticAI, or custom orchestration frameworks
- Research or applied work in reinforcement learning, chain-of-thought reasoning, or graph retrieval
- Understanding of SEC filings or regulated workflows
- Advanced degree in CS, ML/AI, or equivalent depth of experience
Why Join
Competitive cash and equity comp. This is a rare chance to be at the front row seat to define a new category — turning compliance from cost center to data asset.