Staff Backend Engineer- Agent Experience (AX)
Confirmed on the employer's own hiring board on Oct 17, 2026. First seen by Alion on Jun 21, 2026.
Cloudinary is the media experience platform trusted by millions of developers and thousands of companies - from fast-growing startups to the world's largest brands - to power the images and video behind their digital experiences.
Something fundamental is changing about who uses that platform. Increasingly, the "developer" calling our APIs isn't a person - it's an agent: an LLM, a coding assistant, an autonomous pipeline acting on a human's behalf.
Our Agent Experience (AX) R&D team exists to make Cloudinary the default media layer those agents run on. Our guiding principle is simple: make agentic workflows work flawlessly with no friction.
We're a small, fast, high-ownership team, and we're looking for a Senior Backend Engineer to make Cloudinary work for agents - both by building new agent-facing infrastructure and by reshaping our existing platform for a world where the caller is a machine. Some of this is greenfield. Much of it is taking APIs, auth, and SDKs and making them ergonomic, legible, and reliable for agents.
It’s a problem the whole industry is still figuring out. You’ll help define what great API design looks like when the caller is not a human developer, but an agent acting on their behalf.
What you'll do
- Build the surfaces agents call directly: API-first account provisioning, machine-to-machine authentication, scoped credentials, and agent-level identity and attribution.
- Own our MCP (Model Context Protocol) servers and an agent-first CLI - let coding agents and LLMs discover and invoke Cloudinary reliably, engineered for agentic callers: token-efficient tool schemas, machine-parseable output, and loud, actionable failures instead of silent ones.
- Harden the agent media path - make upload and core media operations reliable for non-human callers, eliminating the ambiguity and silent-failure modes that cause agents to drop off mid-task.
- Make agent behavior measurable - partner on the instrumentation, signals, and attribution that tell us whether agents are succeeding.
- Work across a polyglot backend - Node.js/TypeScript for the agent tooling layer, with Ruby/Rails and Go in the core platform.
- Ship, measure, iterate - prototype quickly, dogfood the experience, and represent the agent's perspective in every design decision. Work closely with Product, Ecosystem, and Discoverability across the AX team.
About you
10+ years building and operating production APIs, SDKs and Backend systems and at scale.
Strong in Node.js/TypeScript, and comfortable working in (or quickly ramping on) Ruby/Rails and/or Go.
Deep API and protocol design fundamentals - authentication (OAuth, API keys, scoped permissions), versioning, idempotency, rate limiting, clean contracts.
Genuine curiosity about agents and LLMs and how they're reshaping how software gets built and run.
Solid distributed-systems and cloud instincts (AWS or equivalent).
A bias for ownership in ambiguity - you scope the problem, not just the solution, and you ship.
Strong written communication - it matters more than usual when your consumers are machines and your teammates are distributed.
Hands-on experience with MCP, LLM tool/function-calling, or building AI-/agent-native products.
Experience building SDKs, CLIs, or developer tools that other engineers depend on.
Background in media, imaging, or video processing/delivery.
Open-source contributions in the agent or developer-tooling ecosystem.