Agent Systems Engineer, Finite State

• Strong engineering background with production code, APIs, auth, and data flow fluency
• Hands-on experience building with LLMs and agent frameworks: tool use, retrieval, evals, prompt engineering, MCP, current-generation agent SDKs
• Deep familiarity with the modern GTM stack: Salesforce or HubSpot at minimum, plus working knowledge of outbound sequencers, conversation intelligence, enrichment platforms, and product analytics
• Process thinking: you map full workflows, edge cases, and human handoffs, and you have opinions about what should stay human
• Self-directed on the business side: you can run discovery with a VP of Sales, define the process yourself, and ship a working v1 without a PM translating
• Judgment about revenue work: you can tell the difference between something that drives pipeline and something that just looks good in a dashboard
• Strong sense for security, governance, and risk
• Comfort with ambiguity and broad cross-functional scope

Desirable

• This is an internal Forward Deployed Engineer role with a GTM focus. You’ll embed with Sales, Marketing, RevOps, and Customer Success to learn how revenue actually moves through Finite State, then rebuild those workflows as governed AI agents wired into the tools your stakeholders already use
• You’ll be first in the seat, so you own both halves of the job: the process side (sitting with revenue leaders, mapping how the work really happens) and the build side (writing the code, wiring the systems, standing up the agents)
• Embed with revenue leaders across Sales, Marketing, RevOps, and CS to map workflows end-to-end, including the messy parts and human handoffs
• Design and build AI agents for revenue-critical workflows: ICP research, account scoring, signal-based plays, outbound personalization, CRM hygiene, deal intelligence, churn risk, expansion triggers, lead routing
• Integrate agents with our GTM stack via APIs, MCP servers, and custom connectors: Salesforce, HubSpot, Clay, LinkedIn Sales Navigator, Slack, our own product telemetry amongst other tools
• Build signal pipelines that capture buying intent (hiring patterns, funding events, security disclosures, product usage) and trigger the right action automatically
• Codify recurring revenue motions as reusable skills so the next agent doesn’t start from scratch
• Stand up the governance layer: permissions, audit trails, access controls, sensitive data handling, and rollback paths
• Build evaluation harnesses that measure real business outcomes (pipeline generated, deals accelerated, rep hours saved), not just whether the agent ran
• Expand into adjacent internal functions (Finance, People, Security ops) as the pattern proves out