What are the responsibilities and job description for the AI Engineer San Francisco position at Arceus?
About
Arceus is building the AI-native law firm for the next generation of category-defining companies. We help fast-moving companies move contracts through legal at record speed, with transparency, predictable pricing, and top quality.
We're hiring a Founding AI Engineer to build the intelligence at the core of our platform. You'll work directly with attorneys to encode real legal judgment into AI systems that get faster and more accurate over time.
THE OPPORTUNITY
We are seeing incredible demand for our solutions. The market is open and the window is short. The category will be defined in the next 12 to 18 months, and we intend to win it.
This Is a Rare Chance To
You build AI systems that work in the real world, not just in demos. You know that the hard part isn't calling a model — it's evaluation, retrieval, prompt and pipeline design, and the relentless work of making outputs trustworthy. You're excited to sit with attorneys, understand how they think, and turn that into systems that hold up under real scrutiny.
You're rigorous about quality and obsessed with measuring it. You move fast, but you respect that in this domain, a wrong answer delivered confidently is worse than no answer.
What Success Looks Like
Arceus is building the AI-native law firm for the next generation of category-defining companies. We help fast-moving companies move contracts through legal at record speed, with transparency, predictable pricing, and top quality.
We're hiring a Founding AI Engineer to build the intelligence at the core of our platform. You'll work directly with attorneys to encode real legal judgment into AI systems that get faster and more accurate over time.
THE OPPORTUNITY
We are seeing incredible demand for our solutions. The market is open and the window is short. The category will be defined in the next 12 to 18 months, and we intend to win it.
This Is a Rare Chance To
- Define what AI-native legal work actually looks like, technically
- Own meaningful equity in a company reshaping a trillion-dollar market
- Turn expert human judgment into a defensible product moat
- Build at the frontier of applied AI on problems where accuracy genuinely matters
You build AI systems that work in the real world, not just in demos. You know that the hard part isn't calling a model — it's evaluation, retrieval, prompt and pipeline design, and the relentless work of making outputs trustworthy. You're excited to sit with attorneys, understand how they think, and turn that into systems that hold up under real scrutiny.
You're rigorous about quality and obsessed with measuring it. You move fast, but you respect that in this domain, a wrong answer delivered confidently is worse than no answer.
What Success Looks Like
- First 30 Days: You've shipped AI features into the product and stood up the evaluation harness we use to trust them.
- Months 3–6: You own core AI pipelines, accuracy is measurably improving, and review playbooks are feeding directly into the system.
- Year 1: The platform's intelligence is a real moat — built on systems you designed — and you've helped grow the AI team and its standards.
- Design, build, and own the AI systems at the core of the platform: retrieval, extraction, generation, and the pipelines around them.
- Build rigorous evaluation so we can measure and trust output quality, and improve it over time.
- Partner directly with attorneys to translate review playbooks and judgment into AI systems that get better with use.
- Work with software engineers to ship AI features reliably into production.
- Stay close to the frontier and make sound calls on where to use which models, tools, and techniques.
- Own accuracy and reliability — in this domain, quality is non-negotiable, and customer experience always comes first.
- 4 years in software/ML engineering, with hands-on experience building LLM-powered or ML products in production
- Strong grasp of applied LLM techniques: retrieval (RAG), evaluation, prompt and pipeline design, fine-tuning where appropriate
- Track record of shipping AI features that perform reliably with real users and data
- Strong software fundamentals and comfort owning systems end to end
- Comfort with ambiguity and a fast-moving, high-ownership environment
- Bonus: experience in legal, fintech, or other domains where accuracy is high-stakes
- Excellent judgment, communication, and a high bar for quality
- Competitive base salary commensurate with experience: $180,000 – $260,000
- A top-of-market equity stake in the company
- Health, dental, and vision coverage; unlimited PTO
Salary : $180,000 - $260,000