AI can write the code now. The hard part is knowing exactly what to build.

I'm working with a fast-growing, 100% remote life sciences SaaS company that is building a new, AI-first version of its core product.

Almost all of the code is generated by AI, so the engineers focus on modelling the domain, writing clear specs, making design decision and rigorously reviewing what gets built.

It's proper engineering, just earlier in the process.

You'd work in a pod of two engineers with a product manager, as part of a product, engineering and QA team of 20+.

• Why it's interestingThey're at the cutting edge, of spec-driven, AI-first development and you'll help shape how it's done.
• It's a brand-new product that will replace the existing system
• You'd own features from the first conversation with product through to delivery, and propose design decisions along the way
• The work matters, because their customers make life-saving and life-enhancing products

• What you'll need7+ years hands-on backend, ideally Go (or Python/TypeScript)
• Requirements engineering: turning business needs into clear, testable specs
• You've designed a significant system, or subsystem from a domain model you built with business experts, using DDD or strong enterprise design patterns
• Microservices and REST API design
• Fully embracing AI: daily use of tools like Claude Code, Cursor or Copilot to generate code from specs
• Product sense and clear communication
• SQL, MongoDB or Elasticsearch, and Kafka, SQS or MQTT
• Docker, basic Kubernetes, AWS/Azure/GCP, CI/CD and solid testing