SAGARIS is one brain across CRM, calls, texts and social, staffed by agents you govern. The Company Brain is the system underneath that: it ingests every revenue signal a team produces, models the state of the pipeline, and gives agents a trustworthy place to act from.
As a founding engineer on Agents & Infrastructure, you own that layer end to end: the runtime that executes agent work, the data plane that feeds it, and the guardrails that make autonomy safe to ship. You will set patterns the rest of engineering builds on for years.
What you'll do
- Design and build the agent runtime: planning, tool use, retries, and the approval gates that keep a human in the loop exactly where it matters.
- Own the data infrastructure that unifies CRM, calls, texts and social into a single model of the pipeline.
- Make autonomy auditable: every draft, send, and field an agent touches is attributable and reversible.
- Set the reliability bar (latency, evals, and observability) for systems that act on a customer's behalf.
- Partner with forward-deployed engineers to turn field learnings into durable platform capabilities.
What we're looking for
- You've built and operated production backend systems at scale: data-intensive, low-latency, or agentic.
- Deep comfort with distributed systems, queues, and the failure modes of software that takes real-world actions.
- You treat correctness and observability as features, not afterthoughts.
- You've worked close to ambiguity and shipped anyway: founding, early-stage, or 0-1 team experience.
- Fluency in a modern backend stack (we use TypeScript and Python; strong engineers pick these up fast).
Nice to have
- Hands-on experience wiring LLMs into production: tool calling, evals, retrieval, or agent frameworks.
- Familiarity with the revenue/GTM data landscape (CRMs, dialers, messaging, enrichment).
- A track record of setting engineering patterns others adopt.
What success looks like
- In 3 months: the agent runtime executes multi-step work reliably, with approval gates and a complete audit trail.
- In 6 months: the data plane unifies at least three signal sources into one model design partners trust.
- In 12 months: teams turn agents up because the reliability and auditability have earned it.
Benefits
- Meaningful founding equity, with a real stake in what you help build.
- Top-of-band cash compensation. We don't ask you to trade salary for story.
- Remote-first with deliberate in-person offsites; we optimise for deep work, not attendance.
- A hardware and home-office budget, plus whatever software makes you faster.
- Comprehensive health coverage and a genuinely respected time-off policy.
- Direct access to the founders and an outsized say in the roadmap and the culture.