About Plum
Plum is re-imagining employee healthcare & insurance benefits for fast-growing Indian businesses. We combine modern insurance products, primary, preventive care and data-driven claims to protect 5,000+ companies and 1 million+ lives today. Our next milestone—10 million lives by 2025—demands that every team can move faster and smarter than ever.
The Team & Role
We’re hiring an AI Automation Engineer to lead how we apply AI inside Plum. Your mandate is simple: turn manual, repetitive work into intelligent, self-serve systems. You will partner with Claims Ops, Sales, Marketing, Product, Finance and Customer Success to uncover the highest-leverage problems and ship LLM-powered tools.
The best person for this role doesn't wait to be handed a spec. They walk into a team, understand the business pain, and come back with a working prototype. You'll be as comfortable in a business review as you are in a codebase.
What You’ll Do
- Discover: Shadow teams and run quick sprints to identify automation opportunities.
- Build: Prototype and ship AI agents, chat interfaces, and micro-apps—then iterate fast.
- Use AI to build AI: Code-gen, auto-testing, eval harnesses.
- Full-stack ownership: From prompt engineering & RAG → to React/Next.js front-ends → to serverless deployments.
- Add guardrails: Privacy, feedback loops, bias checks so tools are trusted.
- Enable adoption: Demo, document, and measure ROI to get teams onboard.
What We’re Looking For
Technical
- Strong skills in modern web development: JavaScript/TypeScript, React/Next.js
- Comfort with APIs, databases, and shipping real features end-to-end
- Curiosity about AI-first workflows like Cursor, advanced prompting, RAG, and evals
- Proof of work: projects, demos, repos, or a write-up of something you've built
Business and Product Mindset
- You think in outcomes, not outputs. You care whether the thing you built actually moved a metric
- You can sit with a Claims Ops lead or a Sales manager, understand their bottlenecks, and translate that into a technical solution without needing a PRD written for you
- You have a nose for leverage. You instinctively prioritise the automation that saves 500 hours over the one that saves 5
- You communicate clearly across functions: to engineers about architecture, to business teams about ROI, to leadership about what's next
- You treat every tool you ship like a product, with users, feedback loops, and iteration cycles
Bonus Points
- Experience with Python for backend workflows
- Familiarity with vector DBs, LangChain/LlamaIndex, or eval frameworks
- Open-source contributions or blog posts on LLM/AI development
- You've built in insurance, fintech, or ops-heavy domains
- You've run your own evals or A/B tests on LLM outputs and drawn real conclusions from them
Proof of Work Matters
We care more about what you’ve built than your CV. Show us:
- Links to repos, demos, or projects (side projects count!).
- If closed-source, share a short write-up or demo video.
A strong application = working code > screenshots > slides.