Grove Collaborative is a sustainability-focused consumer products company creating household and personal care essentials that are effective, beautifully designed, and healthier for people and the planet. We are a certified B Corp, plastic-neutral, and on a mission to transform the CPG industry for good.

THE ROLE

Grove is hiring an AI-First Android Engineer — a senior, single-threaded owner of our Android surface who operates as an “engineer-as-a-team.” This isn’t a role where you hand-write every line and lean on AI for autocomplete. This is a role where AI is your primary instrument: you direct fleets of coding agents, generate, review, and refactor at orders of magnitude beyond what a traditional engineer can produce, and take total ownership of features end-to-end — from discovery through production. You will be measured by what ships, what stays up, and what customers feel — not by lines of code or tokens used.

You operate in a new-world environment where one well-equipped engineer can deliver the output of a small team. We expect you to compound your own leverage every quarter: better prompts, better evals, better agent harnesses, better guardrails. The bar isn’t “uses AI tools.” The bar is “runs AI like a team you manage.”

WHAT YOU’LL OWN

Direct multiple AI coding agents in parallel to design, build, test, and ship Android features in Kotlin against the Android SDK.
Take total ownership of a slice of the Android app — discovery, scoping, architecture, implementation, QA, release, monitoring, and iteration.
Run an “engineer-as-a-team” workflow — spawn agents for spec drafting, codegen, test generation, code review, refactoring, doc writing, and bug triage, then merge, judge, and ship.
Develop and defend a strong point of view on what good Android code looks like inside Grove’s architecture (Clean Architecture, multi-module Gradle with base/middleware/data/app, MVVM with DataBinding, Koin for DI, Coroutines/Flow, Moshi) so you can critically review AI output and reject what doesn’t meet our bar.
Design prompts, evals, and harnesses that make AI-assisted mobile development repeatable and measurable, not artisanal.
Stand up automation that closes the loop: Fastlane lanes, CircleCI checks, AppCenter/Play Console distribution, device test farms, telemetry-driven rollback, and AI-powered review gates.
Drive architecture decisions with AI as a research and prototyping partner — explore three options in a day instead of one in a week, including the path to introducing Jetpack Compose into a DataBinding/XML codebase.
Collaborate with other engineers on architecture and trade-offs — bring designs for review before you ship them. Solicit business context from PMs, designers, and platform engineers; don’t guess in isolation.
Own production health autonomously — monitor Sentry, Datadog, Cloudflare, and Firebase to assess system health, triage incidents, and re-prioritize your own backlog without being asked. You set the queue; you don’t wait for someone to tell you what’s broken.
Mentor the Heady contractor team on AI-native workflows and share learnings with mobile peers.
Triage Sentry crashes, hotfix, and root-cause incidents with AI-assisted investigation and remediation (we already run Sentry’s auto-fix agent — you’ll push it further).

ABOUT YOU

5–8 years of professional Android development experience with a portfolio of shipped consumer apps at scale.
Native, demonstrable fluency with modern AI coding tooling (Claude Code, Cursor, Copilot, Codex, agentic workflows) — you can walk us through a feature you shipped where AI did the majority of the coding and you did the orchestration.
Strong, opinionated taste in Kotlin and modern Android architecture (Clean Architecture, MVVM, multi-module Gradle, Koin for DI, Coroutines/Flow, Moshi) — and a clear point of view on when MVI or Compose is worth introducing.
Comfort working in a DataBinding/XML codebase today, with the judgment to drive a phased Compose migration when and where it’s worth it (not a rewrite-from-scratch reflex).
Experience integrating with REST + JSON:API-style backends and the Shopify Buy/Checkout SDKs at consumer scale.
A bias toward owning end-to-end outcomes rather than handing off work at team boundaries — paired with the judgment to know when to pull another engineer in for an architecture review or business-context check.
Genuinely autonomous in operation: you read Sentry/Datadog/Firebase/Cloudflare dashboards on your own, prioritize your own queue, and don’t need a manager to surface what’s on fire.
Comfort writing prompts, evals, and guardrails — you treat AI as a system to be engineered, not a search box.
Sharp written communication — specify clearly, review rigorously, and explain decis

AI-First Android Engineer at Grove Collaborative