Senior Software Engineer II, Agentic AI

DigitalOcean•Seattle, WA

1d•$167,000 - $209,000•Hybrid

About The Position

DigitalOcean is building infrastructure that enables developers to build, run, and scale AI-powered applications and autonomous coding agents. As a Senior Software Engineer on the MARS (Managed Agent Runtime Services) team, you will help build a production-grade platform for securely, reliably, and scalably running coding agents and agentic workflows. You will design and implement core managed-agent-runtime components, including agent-framework integrations, execution environments, control-plane APIs, event-driven infrastructure, and developer-facing APIs and tools. The work spans distributed systems, cloud infrastructure, and emerging agentic AI technologies, with the goal of making it easier for developers to build sophisticated agents without managing the underlying infrastructure. The role calls for strong software engineering fundamentals, pragmatic architectural judgment, end-to-end ownership, and a focus on simplifying complex technology for developers.

Requirements

8+ years of professional software engineering experience.
Experience working with microservices, messaging systems, databases, and infrastructure as code, with a strong emphasis on large-scale SaaS or PaaS products.
Experience with Go or similar backend languages.
Hands-on experience integrating with LLMs or multimodal models and creating, managing, and operating agentic workloads for high-scale customers, or working knowledge of complex multi-day agent session lifecycles, pause/resume transitions, live token-stream fan-out using Kafka and SSE, and sub-second human-in-the-loop approvals and state transitions.
Proven experience building high-throughput REST/gRPC APIs, custom reverse proxies, or network routers that handle concurrent connection scaling.
Practical experience with cluster schedulers such as Kubernetes or Nomad, service meshes, or distributed key-value stores such as Redis, Consul, or Etcd for real-time state tracking.
Strong experience modeling application state in relational databases such as PostgreSQL or MySQL and designing high-volume telemetry ingestion pipelines with data stores such as Snowflake or ClickHouse.
Experience building products for developers, such as APIs, SDKs, or cloud consoles, with empathy for developer workflows and latency constraints.
Proven experience shipping customer-facing software products and operating critical services in a high-scale environment similar to DigitalOcean.

Nice To Haves

Familiarity with agentic AI orchestration challenges, LLM frameworks such as LangGraph, OpenAI Agents SDK, or CrewAI, Model Context Protocol (MCP), or human-in-the-loop (HITL) primitives.

Responsibilities

Design, implement, and operate core services for provisioning, executing, managing, and monitoring AI coding agents and agentic workloads.
Build and maintain integrations with agent frameworks and runtimes such as OpenAI Codex, Claude Code, CrewAI, and LangGraph, as appropriate to the product roadmap.
Develop scalable, fault-tolerant services and APIs for agent sessions, lifecycle management, execution orchestration, and asynchronous event processing.
Collaborate on sandbox- and microVM-based execution, including isolation, resource limits, secrets management, network controls, and secure interactions between agents and external services.
Build APIs, CLI integrations, SDK-compatible interfaces, and reusable examples that help developers integrate agents, tools, triggers, and external systems into their applications.
Instrument services and agent executions with logs, metrics, traces, and actionable events; improve debugging, incident response, performance, and service-level objectives.
Own features end to end, from translating product requirements into technical designs and implementation plans through tests, production rollouts, and operational support.
Lead design discussions, conduct code reviews, improve engineering practices, and proactively address technical debt and operational risks.
Partner with product management, infrastructure, security, inference, and developer experience teams to deliver cohesive customer experiences.
Evaluate and prototype agent orchestration, multi-agent workflows, tool execution, human-in-the-loop controls, and usage policies, and turn promising ideas into production-ready capabilities.