Adobe is seeking a Senior Software Development Engineer to lead the workflow layer for Firefly Boards' agentic capabilities. This role involves designing and building workflows, ensuring the reliability and safety of long-running tasks, and contributing to the strategic direction of the product in collaboration with senior engineers.
Responsibilities:
- Design and build the workflows that drive Boards' agentic product. Implement what the product needs, tune agent behavior for production use, and expand the agentic surface as new capabilities come online. This is the engineer who makes the workflows actually work
- Own the gates and guardrails on long-running tasks. Design and build the safety, quality, and cost controls that wrap agent operations: when an agent can act autonomously, when it needs to pause or escalate, how it recovers from partial failure, how budgets and limits are enforced. Most production agent systems fail here; this work is what keeps Boards' from doing the same
- Design for survivability across interruption and resumption. Build agent workflows so multi-step operations can be checkpointed, paused, resumed, and recovered cleanly across the time horizons Boards needs. Consume the contracts the platform exposes for these capabilities; the platform itself is not in scope
- Partner with the Senior Staff engineer on shaping where the agentic product goes next. That engineer sets strategic direction; this role contributes workflow-level depth that informs what's actually feasible
- Contribute to evaluation infrastructure for agent quality, latency, and cost. Not the primary owner, but a meaningful contributor
- Stay current on the frontier of agentic systems and bring back what's worth adopting
Requirements:
- Bachelor's Degree or equivalent experience in Computer Science
- 5+ years of product engineering experience, with significant ownership of production systems
- Demonstrated recent depth in agentic systems, workflow runtimes, or comparable LLM-tool-using applications. Recency matters more than tenure here; the field is young
- Strong backend systems intuition: state management, long-running operations, recovery from partial failure, observability. Has built and operated systems with these properties, not just consumed them
- Proficient in TypeScript and Python
- Track record of writing clean, testable code and contributing to a codebase that other engineers extend
- Strong written and verbal communication. Comfortable shaping technical decisions in a team where senior peers will push back hard
- Experience with workflow orchestration runtimes for tool-using LLM applications (LangGraph or comparable)
- Experience with production LLM systems: evaluation, observability, latency and cost optimization at the workflow level
- Familiarity with MCP (Model Context Protocol) and agent tool design patterns
- Production Kubernetes experience as a consumer of a managed platform
- Experience with real-time collaborative or multiplayer applications
- Track record of using observability and analytics data to drive engineering decisions: instrumenting systems to answer the questions that matter, then acting on what comes back
- Experience embracing AI-augmented engineering workflows and sophisticated orchestration of agents in the development process itself