Cachet is building the insurance infrastructure to let Europe's platform economy thrive. We design and operate smart, data-driven insurance solutions for shared mobility and digital platforms - enabling fairer pricing, flexible coverage, and real-time risk insights.

As our first dedicated AI Operations Engineer, you aren't just joining a team, you are defining how the entire company works. You will transform how Cachet operates internally: replacing manual, repetitive work with intelligent, automated workflows that let our people focus on what actually matters.

The Mission

You will embed across every team at Cachet, understand how we work at the detail level, and build the AI-powered infrastructure that makes us dramatically more efficient. This is equal parts listening, research, and engineering, you'll spend real time with stakeholders, earn their trust, diagnose what's broken or slow, and then ship something that fixes it.

You will do the workflow diagnosis. You will shadow real work, spot inefficiencies, identify broken loops, and turn messy human processes into sharper, AI-augmented systems. You won't be handed a spec, you'll generate it yourself, from the ground up, by understanding what people actually do every day.

On top of that, you'll build and maintain what we call the AI Brain: company-level intelligence systems that help Cachet remember decisions, retrieve context, avoid repeated work, and compound institutional knowledge in the right direction. This isn't just automation, it’s infrastructure for how we think and operate as an organisation.

Reporting directly to the CTO, you'll work within a clear strategic direction and have real autonomy in how you get there. You'll help shape how Cachet thinks about and works with AI internally, with leadership close enough to unblock you and trust you to execute.

Your Tech Stack & Toolkit

• Languages: Expert-level Python; comfortable working across APIs and integration layers
• AI/ML: LLMs in production (OpenAI, Anthropic, or similar); agent frameworks (LangChain, LangGraph, n8n, or equivalent); workflow automation tooling
• Infrastructure: AWS (ECS/EKS, Bedrock), Docker; comfortable building and deploying production-grade tooling
• SaaS integrations: Slack, Notion, Google Workspace, HubSpot, and whatever else teams depend on
• Scope: 50% building internal tools, agents, and automations — 30% stakeholder discovery, research, and tooling evaluation — 20% cross-functional collaboration with Engineering, Ops, and Leadership.
  
  

What Success Looks Like

3 Months: You've done the discovery. You know where the biggest manual bottlenecks are across every team. The internal AI ops stack is defined and the first tools are live.

6 Months: Multiple teams are running on AI-assisted workflows. Measurable hours saved per week. You're known across the company as the person who makes things faster and smarter.

12 Months: AI ops is established as a core pillar of how Cachet runs. You've moved from individual automations to a mature internal platform other teams can build on. You've documented the playbook, demonstrated clear ROI, and started thinking about what comes next.

What We Expect from You

• Mid+ or senior engineer with hands-on experience building with LLMs in production
• Build and maintain company-level intelligence systems that help Cachet remember decisions, retrieve context, avoid repeated work, and compound institutional knowledge in the right direction
• Proven ability to build and ship internal tools, agents, or automations end-to-end
• Comfortable with AWS and containerization; you can own infrastructure, not just code
• Experience integrating with SaaS tooling and internal APIs
• You default to building rather than workarounds and you know when not to build
• As comfortable in a 1:1 with a non-technical colleague as you are in a terminal, you translate between "here's a problem" and "here's what I'll ship"
  
  

The Nice-to-Haves

• Experience in an AI platform, internal tools, or developer productivity role
• Background in Fintech, Insurance, or another regulated industry
• Experience with agent frameworks (LangGraph, CrewAI, OpenClaw) or AWS Bedrock
• Background in a startup or scale-up where you owned architectural decisions
  
  

Recruitment steps

• HR meeting 30min
• CTO tech meeting 1h
• CEO meeting 30 min
• Background check
  
• Offer

The applications will be reviewed on a rolling basis.