Remote Frontend Jobs
Docker logo

Principal Software Engineer, Docker Agents

DockerLondon, United Kingdom
DockerGithubAPIsPrototypingTechnical Direction
Apply now

We are looking for a Principal Software Engineer (Docker Agents) to join Docker’s AI engineering team to build the future of containerized AI agents. Docker containers are the perfect vehicle to host and run AI agents—providing isolation, portability, and reproducibility. You’ll be working on cagent, our open-source project (https://github.com/docker/cagent), and expanding on it to enable developers to build, deploy, and scale intelligent agents using Docker’s container technology. This is a greenfield opportunity to shape how developers leverage containers for AI agents at massive scale. You’ll define the technical vision, lead architecture decisions, and partner with engineers and leaders across Docker to bring containerized agent capabilities into Docker’s developer experience.

Responsibilities

  • Technical Leadership & Architecture: Define and drive the long-term technical strategy for Docker’s containerized agent platform, including core primitives, APIs, and extensibility patterns
  • Build Containerized Agent Systems: Design and implement systems that leverage Docker containers as the ideal runtime for AI agents, ensuring isolation, scalability, and portability
  • Expand cagent: Maintain and evolve the open-source cagent project, adding new capabilities for containerized agent deployment, orchestration, and lifecycle management
  • Agent Runtime Development: Build robust infrastructure for packaging, deploying, and managing agents in containers across local and cloud environments
  • Evaluation & Testing: Define evaluation frameworks to measure agent quality, reliability, and production readiness; plus the deployment effectiveness of containerized runtimes
  • Reliability & Operability: Establish standards for observability, performance, and operational excellence; lead critical production decision-making and incident learnings as needed
  • Rapid Prototyping: Iterate quickly on new agent capabilities and deployment patterns, moving from concept to production efficiently
  • Open Source Community: Engage with the cagent community, review contributions, and help grow the ecosystem
  • Cross-functional Collaboration: Lead cross-functional technical discussions and influence architectural decisions across Docker’s AI initiatives (including sister teams and platform efforts)
  • Mentorship & Enablement: Mentor senior engineers, raise the bar through design reviews, and accelerate team execution through clear technical direction and coaching
Qualifications
  • 10+ years of software engineering experience, including 3+ years in technical leadership roles (Staff/Principal level or equivalent scope)
  • Go Expertise: Strong proficiency in Go (this is absolutely required) - Docker’s primary language for backend systems
  • AI/ML Knowledge: Practical experience with large language models (LLMs) and agent development patterns
  • System Architecture: Proven ability to design scalable, distributed systems in production environments
  • Container Technology: Deep understanding of Docker, containerization best practices, and container orchestration
  • Cloud/Platform Depth: Experience building and operating platform services with strong foundations in observability, CI/CD, and security principles
  • Operational Excellence: Experience operating and evolving high-availability production systems with a focus on reliability and performance
  • Influence & Communication: Exceptional communication skills and ability to influence across technical and business domains
  • AI Frameworks: Experience with CrewAI, AGNO, ADK, LangChain/LangGraph or similar AI orchestration frameworks (preferred)
  • Python Proficiency: Experience with Python for AI prototyping and tooling (preferred)
  • Experience with Kubernetes or container orchestration platforms (preferred)
  • Open source contributions and community engagement (preferred)
  • Experience with agent evaluation, reliability, and observability techniques (preferred)