AI Product Engineer

This is a fully remote position, open to applicants in Germany.

📋 Description

• Develop agents that investigate incidents, addressing anomalies and answering the question, "why is production disrupted?" utilizing ClickStack as their foundational platform.

• Create skills rather than mere prompts. Establish a library of reusable skills that encapsulates our team's debugging processes, root cause analysis, ClickHouse query writing, and incident response, so agents can access the appropriate playbook instead of starting anew.

• Take full ownership of the agent stack from start to finish. This includes context engineering, tool design, evaluations, tracing, and cost management. You will be accountable for the agent's performance in a production environment.

• Enhance ClickStack as an optimal platform for running AI workloads. Construct the MCP servers, SDKs, and integrations that enable customer agents to read telemetry, act upon it, and maintain their own observability.

• Engage in open collaboration. Work with OSS contributors and customers, troubleshoot their issues together, and incorporate your findings back into the product.

• Confront the challenging aspects of the role, including latency, cost considerations, context window limitations, evaluation coverage, and hallucinations based on actual telemetry.

⛳️ Requirements

• A minimum of 5 years of software engineering experience, with 1 to 2 years focused on LLM-powered systems or agents in a production setting.

• Proficient backend development skills in TypeScript/Node.js and/or Python, with a comfort level in both languages, even if one is your primary choice.

• Direct experience in building agents that involve multi-step tool usage, planning, memory management, and error recovery. You have successfully launched these agents and navigated their failure modes.

• Background in designing skills (Markdown-based workflow encodings, Anthropic-style or similar) along with a clear understanding of when to utilize a skill, a tool, or both.

• Familiarity with MCP: including server construction, tool design, and considerations around authorization, scoping, and observability in agentic systems.

• Strong evaluation practices, including golden sets, LLM-as-judge, and regression detection methodologies.

• Proficient in SQL, capable of writing ClickHouse queries with ease.

• Comfortable working with Docker and Kubernetes.

• Active involvement in open source projects and the developer community.

🏝️ Benefits

• Flexible work environment - ClickHouse is a globally distributed company that supports remote work, currently operating in over 20 countries.

• Healthcare - The company contributes towards your healthcare expenses.

• Equity in the company - Every new team member joining us receives stock options.

• Time off - Flexible time off in the US, along with generous entitlements in other countries.

• A $500 home office setup for remote employees.

• Global Gatherings – We recognize the value of in-person connections and provide opportunities to engage with colleagues during company-wide offsites.