Staff Backend Engineer, AI Agent Platform

United States & CanadaRemotefull-timelead

OtherStaff Backend Engineer

0 views0 saves0 applied

Apply Now

Quick Summary

Key Responsibilities

Node and TypeScript on the backend Event-driven microservices with Temporal and RabbitMQ at the centre Kubernetes on AWS Postgres, MongoDB,

Technical Tools

OtherStaff Backend Engineer

About the Role

~1 min read

Our AI Agent Platform team owns the platform behind nearly all of Quo's customer-facing AI agents. That includes Sona, our AI agent for calls and texts, and Quo Agent, the workspace agent we've just launched. Agents are already live with customers, so this isn't basic plumbing. The hard problems are the ones ahead.

As a Staff Software Engineer, you'll be the hands-on technical lead for the team. You'll set the technical direction for how agents are built and run at Quo, and you'll write a lot of the code that gets us there. Your work will shape how every customer experiences AI inside Quo, and how every other team at Quo builds with it.

We're bringing AI work that grew up in separate teams onto one shared platform. This role is central to that, and you'll have real ownership from day one.

Own the architecture of the agent platform: how any team at Quo creates and runs an agent, the shared pool of tools and integrations agents call, the single layer every model call goes through, and the execution engine that turns model output into real API and tool calls.
Build advanced orchestration, from one agent coordinating five or more subagents to long-running agents that work on a task for hours.
Design memory and self-learning, so our agents get better from customer feedback over time.
Drive the scale, performance, reliability and cost of agent execution as usage grows across calls, texts and the workspace.
Make it easy for other teams to build on the platform. You'll be its best advocate, with clear docs, good defaults and a lot of pairing.
Mentor engineers through design reviews, code reviews and hands-on coaching, raising the bar for the whole AI domain.
Stay close to what's happening in LLMs and agent tooling, and turn what you learn into practical roadmap proposals.

Node and TypeScript on the backend
Event-driven microservices with Temporal and RabbitMQ at the centre
Kubernetes on AWS
Postgres, MongoDB, Elasticsearch and Redis
Datadog for observability
Web (React), iOS and Android clients on the other side of the API

Language is not a filter. If you've built the right things in Go, Python, Java or anything else, we'd rather talk to you than to someone who knows Node but hasn't.

You've built and scaled the backend of a product with a lot of real users, and you've led the architecture of business-critical services.
You've shipped LLM-powered systems to production, ideally an agent platform or orchestration layer. If you haven't built an agent platform yet but have built something close, like an ML platform, a workflow engine or real-time communications infrastructure, and you've gone deep on LLMs hands-on, we'd still love to talk.
You're strong in TypeScript and Node, or you're fluent in another backend language and keen to ramp up on ours.
You have high agency. You see what needs doing, pick it up and move it forward without waiting to be asked.
You learn fast and you're plugged into the AI world. You know what's changing in models, agent frameworks and tooling, and you have opinions about it.
You write clear design docs, explain trade-offs well and bring other engineers and teams along with you.
You care about customers. You weigh technical decisions against the experience of a small business owner using our product every day.

Nice to Have

~1 min read

Have run Temporal or another durable workflow engine in production.
Have worked on customer-facing AI products, not only internal developer tooling.
Have founded or been early at a startup.