Member of Technical Staff (Software Engineer, Backend API)

Perplexity; Greater London; On-site; GBP 90,000 - 130,000; Full time

Job summary

Perplexity is seeking engineers to own critical APIs end to end, from RFCs to production rollout, across Go, Rust, and Python. You will shape enterprise-scale surfaces that agents use, including hosted MCP, tool calling, and auth flows, while ensuring robust data handling and security under real load.

You will work at the boundary of Perplexity's infrastructure and the external world, collaborating with Research, Product, GTM, and customers to close gaps between builder needs and platform

Qualifications

• Experience designing and operating distributed, high-load systems.
• Experience with frontier models and tools in production workflows.
• Strong sense for developers and customers; API user needs in mind.

Responsibilities

• Own critical APIs end to end: write RFCs, ship implementations, deploy from staging to production.
• Contribute to product codebases in Go, Rust, Python.
• Build surfaces agents consume: hosted MCP, tool calling, connectors, and auth flows.

Skills

Distributed systems design

Frontier models tooling

Cloud infrastructure (major provider)

Kubernetes

CI/CD

API product craft

Platform operations

Tools

Kubernetes

Job description

Perplexity is looking for engineers who want to put planet-scale search in the hands of every developer and every agent. The API Platform is one of the fastest-growing surfaces of the company, and the team behind it owns all of it. The Search API, the Agent API, agent first SDKs, docs - from the request path to the developer experience. You'll work at the boundary where Perplexity's infrastructure meets the outside world, and you'll shape what that boundary looks like as agents become the dominant consumers of APIs. We build with AI, for people who build with AI. If you spend your days working alongside frontier models and tooling, you already know where today's developer platforms fall short and that instinct is exactly what we're hiring for.

What you'll do

• Own critical APIs end to end: write the RFC, ship the implementation, roll it out from staging through canary to full fleet, and keep iterating after launch.
• Contribute directly to product codebases across Go, Rust, Python.
• Build the surfaces agents consume: hosted MCP, tool calling, connectors, and auth flows that let agents onboard and operate on their own.
• Harden the platform for enterprise use: authentication patterns, data‑handling guarantees, guardrails, and the security posture of a public API under real load.
• Own what you run in production: monitors and dashboards as code, alerts that page you, postmortems you write and act on.
• Partner with Research to evaluate frontier capabilities and turn the promising ones into production API features; work with Product, GTM, and customers directly to find the gaps between what builders need and what the platform offers today.

What we look for

• Experience with designing and operating distributed, high‑load systems.
• Build with frontier models and tools - open source and closed source - every day, and you can walk us through your agentic workflow in detail.
• High autonomy and agency, paired with real product sense: you think about developers and customers, what an API user wants next, and where the frontier is heading.
• Working fluency with the core cloud infrastructure stack: a major cloud provider, Kubernetes, CI/CD.
• Depth on at least one side of the role: API product craft or platform operations, and genuine interest in the other.

Nice to have

These are a bonus, not a requirement.

• You've built your own agent loops or harnesses: the fastest way to stand out.
• You've built developer or API tooling in the LLM ecosystem or in adjacent domains where API schemas evolve fast under real customer load, such as fintech or payments.
• You've contributed to open‑source SDKs, frameworks, or integrations that developers actually use.