Software Engineer - Backend

Resolve AI•San Francisco, CA

1d•Onsite

About The Position

Software maintenance and production troubleshooting have become a massive tax of engineering velocity. Resolve AI is solving this by building a transformative, truly autonomous AI Production Engineer that investigates and fixes complex system issues end-to-end. Our founders (Spiros Xanthos and Mayank Agarwal) are the core creators of OpenTelemetry and led Splunk Observability. They have had 2 successful exits to Splunk and VMware. We’ve raised over $190M from top-tier investors including Lightspeed, Greylock, DST, Unusual Ventures, and individual backers such as Jeff Dean (Chief Scientist, Google DeepMind), Thomas Dohmke (CEO, GitHub), Matt Garman (CEO, AWS), Reid Hoffman (Founder, LinkedIn), and Fei-Fei Li (Professor, Stanford). Joining Resolve AI at this stage of our journey is a once-in-a-lifetime opportunity. You’ve already decided that you want to work at an AI-native company that’s pushing the limits of how engineers work, and now you’re looking for the right one.

Requirements

Strong backend engineer with real experience building and deploying scalable, high-performance services in high-concurrency production environments.
Experience with cloud platforms (especially AWS) and Kubernetes, and a solid grasp of databases and messaging systems for cloud-native applications.
Bias for action. You close the loop quickly to get an initial version in front of people, then refine from there.
Not afraid to get in the weeds. You like digging into technical problems, getting your hands dirty, staring at data, and debugging.
You use AI tools to amplify what you can build, and you're eager to go deeper on LLM-powered systems.

Nice To Haves

prior SRE, on-call, or incident response experience.
familiarity with observability and other developer-facing tool internals.

Responsibilities

Build simulation environments where AI agents can be trained and evaluated on realistic SRE work: debugging, incident response, infra changes, on-call grind.
Work directly with Research Scientists and domain experts to translate real SRE scenarios into reproducible, graded environments that are actually hard.
Own the architecture and implementation of large subsystems, from early design through long-term evolution. Infra, tooling, data pipelines, scoring, whatever the job needs.
Design, build, and scale the backend services and APIs that generate, run, and score scenarios in high-concurrency environments.
Bake reliability, scalability, and performance into every layer of the system, especially where it's distributed.
Close the loop fast. Ship a rough version, watch it run, find where it breaks, refine.
Dig into the weeds on realism. The environments should reflect how production actually fails, not a sanitized version of it.
Use AI tools aggressively across the stack, from code gen to scenario generation to debugging, to move faster than you could alone.