SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Fieldguide is automating and streamlining assurance and audit work for cybersecurity, privacy, and financial audits. The company is backed by Goldman Sachs Alternatives, Bessemer Venture Partners, 8VC, Floodgate, and Y Combinator, and is trusted by over 50 of the top 100 accounting and consulting firms.
You'll join a 0→1 team building agent-driven experiences for SOX control testing and internal audit workflows. This is full-stack work on long-horizon AI agents that ship real audit features alongside practitioners.
Key responsibilities:
- Design and build agent-driven experiences for SOX control testing, balancing autonomy with auditor oversight
- Own complex, end-to-end workstreams across the full stack
- Make long agentic runs trustworthy: improve reliability, repeatability, performance, and observability so evidence produces consistent, defensible conclusions
- Turn agent output into finished audit deliverables, including structured workpapers that adapt to firm templates
- Ship code from day one and ramp quickly into full ownership of significant product areas
- Collaborate daily with product managers, designers, and SOX subject matter experts on domain problems
- Help shape technical direction and contribute to how the team thinks about agentic systems and AI tooling
At Senior level, you may lead complex projects, set technical direction, serve as domain expert in areas like agent orchestration or document processing, and mentor other engineers.
At Staff level, you may drive company-level technical initiatives, set engineering standards, partner with leadership on long-term strategy, and represent Fieldguide externally.
You should be a strong full-stack engineer who writes code daily and wants to continue doing so. You've built with LLMs in production (not just prototypes) and understand what breaks when real users and documents hit an agent. You're comfortable in ambiguous, fast-moving environments, energized by 0→1 work, and have high ownership. You're a genuine team player across the organization—working directly with PM, design, domain experts, and customer-facing teams. You can pick up complex domain concepts quickly (you don't need to know SOX initially, but you internalize it fast). You're already using AI coding tools and curious about where agentic approaches create real leverage.
What excites you: building the agent and product from scratch on the ground floor of the company's biggest bet; real audit stakes where your work directly affects what firms present to clients; customer proximity with design-partner firms using what you ship within days; high trust and autonomy to define plans for ambiguous problems.
REQUIREMENTS
Must-have:
- Shipped LLM-backed product features to production against real users
- Comfortable full-stack, with enough backend depth to work in agent orchestration
- Autonomy working from an ambiguous spec
- Collaborative mindset with ability to work across PM, design, and domain experts
- Willingness to develop Applied AI skillset including writing evals, using LLM observability tools, and being thoughtful about quality
Nice-to-have:
- Startup experience, as a founder or early engineer
- Track record of 0→1 projects taken from nothing to something people use
- Experience working directly with customers and comfort being in the room when they use what you built
- Experience in internal audit, SOX, financial services, accounting, or compliance
- Familiarity with stack: TypeScript, React, GraphQL, Node.js, PostgreSQL, Python, Temporal
- Experience with document processing, PDF/Excel manipulation, or evaluation frameworks for model output
Not a fit if:
- You're not interested in deeply learning about Agents
- You've never built a product to scale
- You need a fully specified ticket to start
- You'd rather not be in the room with customers and domain experts
