Job DescriptionJob DescriptionSan Francisco, CA · Hybrid · Full-timeCompensation: $220K–$315K + equity\nAbout the Company\nOur client is a Series C, AI-native platform that is changing how audit, advisory, and assurance work gets done. Built for regulated environments, the platform pairs practitioners with agentic AI to modernize engagement delivery from planning through close. Big Four and mid-tier firms run on it, including more than half of the top 100 accounting and consulting firms, and the company recently closed a Series C from leading growth and venture investors.\nFounded 2019 · ~150 people · Industry: AI, Finance, B2B Software\nThe Role\nYou'd join the team that owns the long-horizon agents at the core of the client's AI platform, with a primary focus on their Financial Audit product, the company's largest product line and its next growth engine. The work sits at the frontier of AI product development: agent knowledge, evaluation, and driving quality and reliability at scale. You'd take real ownership of how the team measures and improves agent quality, building on AI infrastructure that is already in place. The agents you ship are used by Big Four firms in production.\nWhat you'll be doing\n\nSet the evals strategy and error-analysis practice that defines how the team measures and improves agent quality\nDesign and build agent knowledge and evaluation infrastructure for long-horizon agents\nLead error analysis on agent behavior and turn findings into platform-level reliability fixes\nBuild and harden the backend systems that support agent execution, evaluation, and monitoring at scale\nHelp drive the platform's reliability and quality roadmap alongside the broader AI team\nMentor engineers and raise the bar on eval rigor across the team\n\nTech stack: Language-agnostic; the team builds with coding agents as part of its day-to-day workflow\nRequirements\n\nCurrently operating at a Senior, Staff, or Tech Lead level (or higher); strong founders are also considered\nRoughly 6–15 years of professional software engineering experience\nOne to two years spent building end-to-end agents or production AI platforms, including agentic loops, evals, and error analysis\nHands-on backend work on customer-facing agents that run in production at scale\nSolid background in backend platform, data platform, or full-stack engineering\nComputer science degree from a top-tier program (US or international)\nAble to work hybrid from the San Francisco office, averaging about three days a week on-site\n\nNice to Haves\n\nTime at a recently IPO'd tech company or an AI-native startup\nDistributed systems experience\n\nWhy Join\n\nBuild customer-facing agents for the client's biggest product, used by Big Four firms at scale\nOwn agents, evals, and workflows on top of AI infrastructure that's already built\nWell-capitalized, late-stage company with a recently closed Series C and marquee investors\nWork on open problems in agent reliability that don't have established playbooks yet\nTwice-yearly in-person offsites; the team averages ~3 days a week in office with no set days\n\nDetails\n\nLocation: San Francisco, CA\nWork policy: Hybrid, averaging ~3 days/week in office (team average, no set days)\nCompensation: $220K–$315K + equity (range spans Senior through Staff)\nVisa sponsorship: Open to visa transfers (e.g. OPT, H-1B transfers)\nEmployment type: Full-time\n