Deployment Engineer

Comet is building the development platform for teams who want to ship robust, reliable, and responsible AI applications. Opik, our open source LLM evaluation framework, has quickly become one of the most popular tools in the space. Our experiment management platform is used by data scientists at companies like Uber, Netflix, and Etsy. Tens of thousands of researchers, engineers, academics, and hobbyists use Comet every day to build the future of AI. 

Working at Comet will give you access to the most exciting work being done in all areas of machine learning. Some of the top researchers and companies working on self-driving cars, drug discovery, particle research, diffusion models, and LLMs use Comet every day. Your work has the potential to accelerate the development of some of the most impactful technology in the world, and you will be doing it alongside a team of passionate, caring individuals. If that sounds exciting, Comet is the right place for you.

Comet is backed by more than $63 million in venture capital funding and powers some of the best machine-learning teams in the world, including Netflix, Uber, Etsy, and Mobileye. We are a remote-first company with offices in New York City (USA) and Tel Aviv (Israel).

Comet is looking for an experienced Deployment Engineer (located in the East or Central USA Only) to help our customers successfully deploy, operate, and troubleshoot Comet in their environments.

This is a highly customer-facing engineering role. You will work directly with customers to understand their infrastructure and deployment requirements, lead deployments, troubleshoot complex issues, and drive problems through to resolution.

You will also contribute to the engineering and automation behind our deployment solutions, including Kubernetes, Helm, Terraform, cloud infrastructure, tooling, and documentation.

• →Work directly with customers to understand their infrastructure and deployment requirements.
• →Deploy and maintain Comet across Kubernetes and cloud environments.
• →Troubleshoot complex infrastructure and deployment issues across Kubernetes, networking, cloud services, databases, and application configuration.
• →Own customer issues from investigation through resolution, collaborating with Engineering, Support, Customer Success, and other teams when needed.
• →Develop and improve Helm charts, Terraform, deployment tooling, automation, and documentation.
• →Participate in technical customer calls and explain complex topics clearly.
• →Use AI tools for troubleshooting, development, automation, and investigation while validating results and maintaining engineering judgment.

• Proven customer-facing technical experience is essential.
• Proven experience using AI tools in real engineering workflows, with the ability to critically evaluate and validate their output.
• Strong hands-on experience with Kubernetes, Helm and containerized applications.
• Production experience with cloud infrastructure, preferably AWS.
• Experience with Infrastructure as Code, preferably Terraform.
• Strong Linux, networking, and infrastructure troubleshooting skills.
• Experience troubleshooting production systems using logs, metrics, and monitoring/observability tools.
• Excellent written and verbal communication skills.

• You take ownership and proactively move problems forward.
• You are passionate about troubleshooting unfamiliar systems and looking for root causes rather than quick workarounds.
• You can work independently while knowing when to collaborate or ask for help.
• You communicate progress, blockers, and decisions clearly - especially in a remote environment.
• You are comfortable with ambiguity and changing priorities.
• You work well across teams and care about both the customer experience and engineering quality.

• Enterprise or self-hosted software experience.
• Multi-cloud or on-premises deployment experience.
• CI/CD and GitOps experience.
• MLOps or AI infrastructure experience.
• Software development experience.
• Startup experience.