Full-Stack Engineer

You own features end to end: the data model, the services, the agent tooling and the screens. On a given week you might wire a new weather dataset into the signal library, give Kite a new tool for structuring a drought cover, harden the rolling-window trigger logic that decides whether a policy pays out, and build the review screen an underwriter uses to approve it.

The core of the platform is agentic. Kite plans, calls tools, runs analysis on client data and produces regulated outputs. Making that reliable — evaluations, guardrails, observability, graceful failure, a human approval step that never gets skipped — is as much the job as any CRUD endpoint.

You will work directly with the CTO and have a large say in architecture, stack and how we build. The platform is growing fast and the next 12 months will set its shape for years.

What you'll do

• Design, build and run backend services for WeatherWise Studio and WeatherWise Construction: APIs, data pipelines, pricing and simulation engines, integrations.
• Build and extend the agent layer — tools, orchestration, memory, evaluation harnesses and the observability needed to trust an agent in a regulated workflow.
• Own the weather data engine: ingestion from 180k+ stations and gridded datasets (ERA5 and others), storage, indexing and fast querying of decades of history.
• Ship the front end of your own features in React/TypeScript to a high standard, working with our design-led front-end engineer on the system.
• Build the embedded API strategy: clean, well-documented, versioned APIs that let brokers and carriers run WeatherWise inside their own systems.
• Own reliability and security: CI/CD, infrastructure as code, monitoring, incident response, and the controls a regulated firm needs.
• Use AI tooling as a first-class part of how you work, and help the whole team get more from it.

What we're looking for

You must have

• 6+ years building and running production systems, with real ownership of backend, infrastructure and front end.
• Strong Python and TypeScript. We care about fluency across the stack more than any single framework, but you should be able to ship a clean React screen as readily as a well-designed service.
• Hands-on experience building LLM-powered or agentic systems in production: tool use, orchestration, prompt and context design, evaluation, and the failure modes that come with them.
• Solid data engineering: pipelines, time-series and geospatial data at scale, and the judgement to pick the right store and query pattern.
• Cloud and DevOps competence (AWS or GCP, containers, infrastructure as code, CI/CD, observability).
• Daily, serious use of AI coding tools (Claude Code, Cursor or similar) and a view on how to work well with them.
• Comfort in a small, senior team: you scope ambiguous problems yourself, make pragmatic calls and ship.

It would be great if you also have

• Experience with weather, climate or geospatial data (NetCDF/Zarr, ERA5, gridded reanalysis, station networks).
• Worked on pricing, simulation or risk models, or in insurance, fintech or another regulated domain.
• Designed public APIs used by third parties.
• Built evaluation frameworks for LLM applications.

We care far more about what you have shipped than which exact tools you used.

Company Benefits

• Fully remote
• 4 day work week
• Plenty of learning and career progression