About The Role

Voice AI that resolves, not transfers.
Most phone systems trap callers in menus and scripts. Vapi is the platform for deploying voice agents that know your business and can listen, adapt, and resolve in minutes.

The numbers: 1 billion calls. 1 million developers. 10x enterprise ARR growth
The customers: Amazon Ring, ServiceTitan, New York Life, Intuit, Kavak, and thousands more, from YC startups to the Fortune 500
The news: a $50M Series B led by Peak XV Partners, with Bessemer Venture Partners, Kleiner Perkins, M12 (Microsoft's Venture Fund), Y Combinator, and our earlier backers. Total raised: $72M

Voice AI that resolves, not transfers
Powering 1 billion calls for companies like Amazon Ring, Intuit, ServiceTitan, and New York Life, Uber, NHS
Trusted by 1 million developers building the future of voice agents
Backed by Peak XV, Bessemer, Kleiner Perkins, M12, Y Combinator, and more with $72M raised
Try talking to Vapi now! https://vapi.ai/

Vapi engineers are already building agent skills, workflows, and tools. We need a dedicated owner to turn these fragmented pieces into one reliable agentic development lifecycle.
We have an opportunity to connect sandboxes, CI, verification, and background agents into a cohesive system that compounds across every engineering team.
We want to ship 10× faster by building the frontier agent harness for real-time voice AI. Agents will place real calls, validate their work against Vapi-specific systems, and ship with confidence.

30 Days: Map and instrument Vapi’s path from task to production, identify the largest constraints on agent effectiveness, and ship your first improvements to our sandbox environment, developer tooling, or validation infrastructure.
60 Days: Launch the first production workflow in Vapi’s internal software factory. Give background agents the context, tools, sandbox, telemetry, and automated quality gates required to turn scoped tasks into review-ready PRs backed by real evidence.
90 Days: Own and launch Vapi’s platform for autonomous, background engineering. Bring the ambition of Ramp Inspect https://builders.ramp.com/post/why-we-built-our-background-agent and Stripe Minions https://stripe.dev/blog/minions-stripes-one-shot-end-to-end-coding-agents to the technically novel and difficult frontier of real-time voice AI. Build a harness that lets agents place real calls, measure latency, conversation quality, inspect telemetry, and iterate to a verified result. Create the foundation that accelerates every engineer and puts Vapi on a credible path to shipping 10× faster.

You have built internal platforms, developer tooling, or infrastructure that materially changed how engineers build, test, and ship software.
You are a hands-on engineer with strong software fundamentals across distributed systems, platform architecture, API design, CI/CD, and monorepo or large-codebase workflows.
You have practical experience building agent harnesses and using coding agents such as Claude Code, Codex, Cursor, or open-source equivalents on real engineering work.
You know how to design secure agent interfaces, MCP integrations, and sandboxed runtimes where agents can call APIs, run CLI commands, and use general-purpose tools safely.
You have built or operated online and offline LLM evaluation systems and can reason clearly about accuracy, latency, reliability, and cost.
You are fluent in at least one of TypeScript, Python, or Go and comfortable working across containers, databases, queues, orchestration systems, and modern cloud infrastructure.
You stay current on frontier models and techniques such as prompt caching, context compression, retrieval, and multi-agent coordination, and you know when the newest tool is useful versus distracting.
You thrive in ambiguity, seek raw input, and turn fragmented experiments into simple systems that other engineers want to use.

Build something worthy of love: Craft matters. We build products and experiences customers genuinely love, not just tolerate.
Commit and follow through: We finish what we start and build trust by being people others can count on.
Why not today? We value urgency and momentum. The fastest path to customer value usually wins.
Seek raw input: We go directly to customers, data, and teammates instead of relying on summaries or assumptions.
It’s our problem: We operate as one team. We share credit, own mistakes together, and support each other when things get hard.
Be direct and kind: We give feedback clearly, respectfully, and without delay.