Software Engineer, Agents (Foundation Agents)

Fieldguide fieldguide.io Open · verified Oct 24, 2026

Hybrid San Francisco Full-time Mid-level Software Engineering

At a glance

Mid-level Software Engineering role at Fieldguide. San Francisco · full-time · $190,000–$220,000 base.

SalaryStated by Fieldguide

$190,000 – $220,000

Base salary for this role, in USD per year, as published in the posting.

This role, from FieldguideOur estimate for the market: $169,488–$240,000Median $200,000

The range includes the market median of $200,000 for mid-level software engineering roles in San Francisco.

Worked out from 212 postings that stated pay, median $200,000. Last updated Oct 24, 2026.

RoleSoftware Engineering

SeniorityMid-level

LocationSan Francisco

WorkplaceHybrid

EmploymentFull-time

Base pay$190,000–$220,000

PostedSep 16, 2026 · 5w ago

Role brief

techroles summary, based on the employer's posting.

What you'll do

Build and maintain the knowledge and evaluation infrastructure used by long-horizon agents
Develop newer agents for customer problems that require more complex work
Study agent failures and use the findings to improve quality
Create backend services for agent execution, evaluation and monitoring
Work with senior engineers on plans to make the platform more reliable and improve agent quality

What you bring

You have hands-on experience creating AI products, rather than only connecting to an LLM API
You understand evaluation methods and how to analyze agent errors
You have backend engineering experience and value dependable systems

Who this fits

This mid-level role suits a backend engineer interested in production AI agents, evaluation and reliability. It is hybrid in San Francisco, with twice-yearly in-person offsites across the U.S.

From the employer

About Us

Fieldguide is establishing a new state of trust for global commerce and capital markets by automating and streamlining the work of assurance and audit practitioners—specifically in cybersecurity, privacy, and financial audits. We build software for the people who enable trust between businesses.

We're based in San Francisco, CA, and backed by Goldman Sachs Alternatives, Bessemer Venture Partners, 8VC, Floodgate, Y Combinator, and more. Over 50 of the top 100 accounting and consulting firms trust Fieldguide to power mission-critical work.

About the Role

The Foundation Agents team stewards the long-horizon agents powering the Fieldguide AI platform. We work at the frontier of AI product development: agent knowledge, evaluations, and improving quality and reliability at scale. If you're excited by long-horizon agents that do real-world work, this is the team for you.

What You’ll Own

Build and maintain agent knowledge and evaluation infrastructure for Fieldguide's long-horizon agents
Build our next generation of agents at Fieldguide, solving increasingly complex customer use cases
Perform error analysis on agent behavior and help turn findings into concrete quality improvements
Build backend systems that support agent execution, evaluation, and monitoring
Partner with senior engineers on the team to execute the platform's reliability and quality roadmap

Who You Are

You've built AI products, not just called an LLM API
You're curious about evals and error analysis. You want to understand why an agent failed, not just that it did
You're comfortable in the backend, with a bias toward reliable systems
You're motivated by long-horizon agents that do real work in production

Experience

Must-have:

Hands-on experience building AI products
Working knowledge of evals and error analysis
Backend engineering experience

Nice-to-have:

Platform engineering skills
Frontend experience
Distributed systems experience

What Should Excite You

Long-horizon agents: working on agents that do real, sustained work and not one-shot demos
Evaluation as a craft: Evals and error analysis are core to how this team improves quality
Platform-level impact: Your work shapes the reliability and quality of every agent built on top of it
Frontier problems: You're working on open problems in agent reliability that don't yet have established playbooks