Enterprises are rapidly deploying agents to automate complex work, and every one of them needs a reliable way to simulate, verify, and optimize those agents before trusting them with real work. Litmus gives them that confidence. We learn how a company's people actually work, generate the sandboxes, tasks, and verifiers to match, and evaluate its agents before they ship and once they're in production. The goal is an autonomous research engine that gives every enterprise the kind of internal benchmarking function the frontier labs have.
Litmus was founded by Sidharth Srinivasan, who led AI evals at Box after its CEO recruited him out of Stanford CS. He built the function from inception, running a research engineering team and owning Box's relationships with the frontier labs and data vendors. Litmus has raised $5M from Pear, BoxGroup, Workbench, and Emergence, alongside angels including Aaron Levie (CEO of Box), Arash Ferdowsi (co-founder of Dropbox), and senior leaders at OpenAI, Anthropic, Google, and Mercor. Demand already spans Fortune 15 enterprises, top US banks, AI transformation firms, and AI-native services companies, and we're hiring to keep up.
How we work: in person in FiDi, five days a week. Small team, high ownership, and we care about outcomes, not hours.
Build the runtime that simulates the real world for AI agents.
Every Litmus evaluation runs inside a simulated copy of a company: its tools, its data, and its users. Thousands of agents run against those worlds in parallel, each run isolated, stateful, and replayable, often inside the customer's own cloud. Nothing off the shelf does this. You'll design it and own the architecture from day one.
What you'll work on:
Stateful sandboxes that spin up, reset, fork, and replay on demand, at tens of thousands of runs
Orchestrating large fleets of agents, fast and cheap
Deploying into the most demanding enterprise environments: customer clouds, redaction, multi-tenancy, and security review
Pulling production traces from the tools customers already use and turning them into tests
The core architecture, and turning it into a product teams adopt on their own
Why it's exciting:
Infrastructure is the moat here, not a dashboard on top of a database.
What you build decides which customers we can say yes to, including some of the largest enterprises in the world.
The runtime for agent simulation is still being invented, and the same environments that test agents today will train them tomorrow.
You're a strong backend or infrastructure engineer who has been early before
You've built and run production systems, and you're excited to own the parts you haven't done yet
You're AI-pilled: you build with LLMs, work alongside coding agents every day, and have opinions about where agents break
You make strong architectural calls with incomplete information
You typically have 3–10 years of experience
Cloud infrastructure across AWS, GCP, or Azure
Containers, sandboxing, and isolation
Distributed systems, workflow orchestration, and data pipelines
Enterprise security and compliance
Front-end only
Roles where you never made the architectural calls
Careers spent entirely inside very large companies
Founding equity and competitive cash
One of the first hires, working directly with the founder and customers from day one
Ownership of the platform architecture from the start
A beautiful office in FiDi
Breakfast, lunch, and dinner
Medical, dental, and vision coverage
Equinox membership or a fitness stipend
Essentially unlimited budget for AI tools and tokens
Relocation support if you're moving to San Francisco
Job details are sourced from the employer's original posting.
Open job postingAbout the company
Litmus provides software transforming machine data into operational insights for Industrial IoT, Industry 4.0, and Edge Computing.