> Markdown version of [/jobs/ext/3953469-backend-engineer-agents](https://www.wearedevelopers.com/jobs/ext/3953469-backend-engineer-agents). 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).

---

# Backend Engineer, Agents

- **Company:** LAPEL, LLC
- **Location:** San Francisco, CA, United States
- **Experience:** Expert
- **Contract:** Permanent contract
- **Skills:** Application Programming Interfaces (APIs), Artificial Intelligence, Databases, Software Debugging, PostgreSQL, Node.Js, Software Engineering, TypeScript, Tailwind, ReactJS, Backend, Agentic-AI, Webpack, Front End Software Development
- **Published:** October 27, 2026
- **Apply:** https://www.careerbuilder.com/job-details/backend-engineer-agents-san-francisco-ca--57fedabc-73f4-449d-bb44-58b1995889ab

## About the Role

Customer Support/Service, Customer Relations, High Tech Industry, Artificial Intelligence (AI), Startup, Memory Hardware, Service Delivery, Debugging Skills, Prototyping, Engineering, * Have 5+ years of backend or software engineering experience, with a track record of building reliable systems in production

 * Have owned large-scale, enterprise products end-to-end
 * Are excited to build the systems that make AI agents more useful, dependable, and easier to improve
 * Are comfortable designing APIs, working with databases, and building workflows that run reliably over time
 * Understand system design, debugging, observability, and production reliability at a practical level
 * Are customer-focused and able to translate product needs into reliable agent infrastructure
 * Enjoy working in fast-paced environments with high ownership and autonomy
 * Are proactive, detail-oriented, and take pride in building high-quality systems

## Description

* Design systems that help agents remember, summarize, and retrieve context across long-running customer interactions
 * Build workflows and tools that reduce failure modes and make agents easier to trust in production
 * Set up the observability and eval infrastructure we use to monitor agent behavior, debug issues, and raise quality over time
 * Help to scale multi-tenant systems that keep customer and agent context isolated, reliable, and ready for enterprise use

Our Stack

Lapel is built end-to-end in TypeScript in a single monorepo.

 * React, Vite, and Tailwind on the frontend;
 * Node.js and Postgres on the backend;
 * Temporal and Pulsar to power orchestration.

We ship quickly, and care about creating a developer experience where anyone can prototype, experiment, and get ideas into production fast., We aim to hire with intention, respect your time, and offer feedback along the way.

 1. Initial conversation (20 minutes) We'll explore your motivations, share what we're building, and answer your questions.
 2. In-depth interview (60 minutes) We'll go deeper into your craft and experience. No live coding. No prep work necessary.
 3. Take-home project (2-4 hours) You'll complete a short, well-scoped exercise that mirrors the kind of work you'd do here.
 4. On-site visit (1-2 days) Join us in the office to review your take-home, grab lunch, and work on something real.