> Markdown version of [/jobs/ext/4625990-agent-platform-engineer](https://www.wearedevelopers.com/jobs/ext/4625990-agent-platform-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Agent Platform Engineer

- **Company:** VARICK PARTNERS LLC
- **Location:** San Francisco, CA, United States
- **Experience:** Experienced
- **Contract:** Permanent contract
- **Skills:** Release Management, Large Language Models, Concurrency, Multi-Cloud, Backend, Kubernetes
- **Published:** November 7, 2026
- **Apply:** https://www.juju.com/job/28_25197_3c491777448019eff39f7ac6c5fad416

## About the Role

* 3+ years building production software with strong backend and distributed-systems fundamentals
 * Instinct for state, concurrency, idempotency, and failure handling: what happens when a dependency disappears mid-request
 * Experience operating production systems at scale (cloud, release management across versions)
 * At least one LLM system shipped to production, with a real grasp of why demoware fails on the tenth run
 * Comfort with fast, low-process development, paired with judgment on rigor

Nice to Have.

 * Durable execution or workflow orchestration (Temporal or similar)
 * Kubernetes, infrastructure-as-code, multi-cloud
 * Shipping into environments you don't control (BYOC, on-prem)

What We Offer. Meaningful equity, real ownership, direct access to how the world's largest enterprises actually run, flexible PTO, free lunch and dinner in the office, Ubers home if you're staying late, and monthly team dinners.

Logistics. On-site in our San Francisco Financial District office, 5-6 days a week. Comfortable with startup hours, for startup upside.

## Description

The Role. You build Varick OS, the platform every agent runs on: the runtime, orchestration, retrieval, and deployment tooling that turns a one-off agent into something we can spin up, monitor, and ship across many clients. Decisions you make here are load-bearing for every engagement.

What You'll Do.

 * Build the agent runtime: long-running workflows against enterprise systems that go down mid-run, with durable execution, safe retries, and idempotency
 * Own orchestration, retrieval, and context assembly, plus the human-approval gates that hold up when a step touches real money
 * Build the eval and tracing infrastructure that keeps agent quality measurable across the fleet
 * Work on the model layer: provider-agnostic routing, fallback, and cost and latency budgets
 * Build the deployment tooling that ships the platform into environments we don't fully control