← All roles

Founding SWE / MLE

Salus· Engineering
Apply on company site
Hybrid (San Francisco, CA / New York, NY), Hybrid (Remote (US))$250k - $500k + equityFull-timePosted 1mo ago

About the role

Salus is the runtime enforcement layer for AI agents. The role is a founding engineer position, working directly with the CTO on core product development, customer deployments, and research-oriented improvements in LLMs, agents, and systems. This is an in-person role in San Francisco / NYC, with remote options in the US.

Responsibilities

  • Build the core enforcement engine, intercepting tool calls, evaluating policy deterministically, and returning recoverable results for the agent
  • Deploy Salus into customer codebases and real agent stacks (e.g., LangChain, LangGraph, MCP)
  • Contribute to research efforts, benchmarks, adversarial evaluation, and problems around agent authorization and provenance
  • Own the surface area from day one as a founding engineer

Qualifications

  • Deep with LLMs and agents, understanding tool calling, context, eval, and production breakage points
  • Strong systems engineering skills with low-latency, correctness-critical code
  • Excellent general SWE ability to go from vague problem to working solution quickly
  • Comfort with both product and research, and willingness to tackle hard open problems
  • Fast mover with taste and willingness to be early in a high-impact environment
  • Bonus: Experience with agent infrastructure, security, verification, enterprise deployments, or research papers

Skills

Required:PythonTypeScriptLLMsAI Agents
Last updated Oct 25, 2026. Interested? Apply on the company site.
Salus logo

About Salus

Guardrails to validate your agent's actions before they execute

Your agent processed a refund without looking up the order ID, costing you thousands. You only found out three hours later from a support ticket. Evals, output scoring, and observability can reduce the likelihood of mistakes like these occurring - but there's no solution that inspects and prevents an action as it’s about to execute. Salus does that. We’ve built an API that wraps around your agent and checks its actions at run time, blocking incorrect ones and providing immediate feedback to guide retries. Kevin and Vedant were roommates at Stanford, where they both studied computer science.

Developer ToolsB2BAPIInfrastructureAI

More open roles at Salus