Design and build the orchestration and planning systems that power Dropbox’s agent platform, including task decomposition, scheduling, and error recovery.
Implement real-time execution frameworks that allow multiple agents and tools to coordinate and act reliably at scale.
Develop clean, extensible APIs, data models, and protocols that enable new agent capabilities to plug seamlessly into the platform.
Build and maintain the evaluation platform used to measure agent performance (accuracy, latency, cost, safety, satisfaction) and catch regressions.
Establish and improve observability, testing, and debugging workflows for agentic systems in production.
Collaborate with AI research, infra, and product teams to translate high-level product goals and research insights into robust, production-ready systems.
Contribute to technical roadmaps, design docs, and architecture reviews, helping set direction for agent platform and eval platform evolution.
Requirements
9+ years of professional software engineering experience, including significant experience building production-scale systems or platforms.
Proven experience building or operating orchestration, workflow, or scheduling systems (e.g., task graphs, workflow engines, pipelines).
Strong technical understanding of system design, service reliability, fault tolerance, API design, and observability.
Demonstrated ability to drive complex technical projects end-to-end and work effectively across research, infra, and product teams.
Strong analytical and problem-solving skills with a focus on measurable outcomes (latency, throughput, reliability, cost).
Excellent written and verbal communication skills, with the ability to explain design decisions and influence technical direction across teams.