Software Engineer, AI Agents
In-Office
Posted 8d ago
Job Location
In-Office
Tech Stack
Remote Work Policy
On-site
Categories
AI Agent Engineer
About the job
At Cloudflare, we are building industrial-scale AI agents to directly support customers, moving beyond research to deliver real-world impact from day one. This role involves assembling existing Cloudflare technologies like Workers, Durable Objects, KV, R2, D1, Vectorize, Workers AI, AI Gateway, and the Agent SDK into functional agents that customers will use daily. The primary goal is to ship production agents on the Cloudflare stack, iterating rapidly through building, deploying, and learning. Your code will serve as the primary interface for Cloudflare customers interacting with these AI capabilities.
Responsibilities
- Build agents on Workers utilizing Durable Objects for state management and short-term memory.
- Integrate tools with the Agent SDK, MCP, and function calling capabilities.
- Utilize Vectorize, KV, R2, and D1 for semantic memory, caching, file storage, and configuration.
- Run models through Workers AI and AI Gateway, integrating third-party services as needed.
- Develop evaluation frameworks, guardrails, and audit mechanisms for continuous improvement and tuning.
- Create agents capable of summarizing information, proposing solutions, and escalating issues to humans.
- Expose agent health and performance metrics through transparent dashboards.
- Integrate with queuing systems and webhooks, publishing events via Queues or Pub/Sub.
- Reduce cost per customer interaction and decrease time to first response, validated by data.
- Take full end-to-end ownership of shipped features, including on-call responsibilities with team support.
- Design and maintain robust observability for distributed AI workflows, implementing structured logging and end-to-end tracing.
- Architect security boundaries for agent operations, including secure credential handling and multi-layer approval gates.
Requirements
- Demonstrated success in shipping production systems, evidenced by public repositories or released work.
- Proficiency in TypeScript or Rust for development on Workers, with experience in HTTP, queues, asynchronous programming, and performance optimization.
- Hands-on experience with Durable Objects, KV or R2, and either D1 or PostgreSQL.
- Practical experience with model tooling, including prompt I/O, tool calling, evaluations, and safety checks.
- A strong observability mindset, including experience with logs, traces, metrics, and redlines.
- Experience with agent-to-agent (a2a) or multi-agent frameworks.
- Experience developing LLM evaluation frameworks, including automated scoring systems and CI-integrated quality gates.
- A preference for simple, scalable design principles.
Benefits
- Impact at global scale serving millions of Internet properties.
- Opportunity to build AI directly into an edge platform operating in 300+ cities.
- Career growth opportunities as an early engineer in the India AI team.
- Leadership opportunities and direct influence on Cloudflare's AI roadmap.
- Culture of ownership, autonomy, accountability, and trust.
- Continuous learning and growth through collaboration with cross-functional teams.
- Mentorship and knowledge sharing opportunities.