About agentmama
We build AI agents that audit and optimize e-commerce advertising. Our platform syncs live data from Google Ads, Meta, Merchant Center, and Google Analytics, runs it through a rule engine, LLM-powered audit agents, and our own prediction models, and turns the findings into concrete actions marketers can apply with one click.
We're a small team shipping fast on a modern, strictly-typed stack. You'd be one of a handful of engineers reporting directly to the CTO, which means real ownership: features you design end-to-end, from the Prisma schema to the React component, ship to customers within days.
What you'll do
Design and build features across our TypeScript monorepo — tRPC procedures and NestJS services on the backend, React on the frontend.
Extend our integration workers that sync campaign, product, and analytics data from the Google and Meta APIs via Cloud Pub/Sub pipelines.
Work on the AI audit agent: prompt design, rule evaluation, and the plumbing that turns LLM output into reliable, deduplicated actions — across multiple model providers, not just one.
Help productionize our in-house prediction models: wiring their output into the rule engine and keeping training data and evaluation pipelines healthy.
Evolve the data model in PostgreSQL/Prisma and keep migrations, multi-tenancy, and role-based access clean as the product grows.
Own quality: write Jest tests alongside your code, review PRs, and help keep our CI/CD pipeline to Cloud Run fast and boring.
Our stackBackend
TypeScript · Node.js NestJS + Fastify tRPC · Zod Prisma · PostgreSQL
Frontend
React · Vite End-to-end types via tRPC
Infrastructure
Google Cloud Run · Pub/Sub Docker · GitHub Actions pnpm workspaces · Turborepo
AI & integrations
Multiple LLM providers In-house prediction models Google Ads · Meta Merchant Center · GA4
What we're looking for
5+ years building production web applications, with strong TypeScript on both server and client.
Solid relational database instincts — schema design, migrations, and query performance in PostgreSQL (Prisma experience is a plus).
Experience with event-driven or queue-based architectures (Pub/Sub, SQS, Kafka, or similar) and the failure modes that come with them.
A habit of testing: you write tests because they let you move faster, not because someone told you to.
Pragmatic product sense — you ask why before how, and you'd rather ship a focused feature than a perfect abstraction.
Nice to have
Hands-on work with LLM APIs in production — prompting, evals, structured output, cost control, and switching between providers.
Applied ML experience: training, evaluating, or serving prediction models alongside product code.
Familiarity with the Google Ads, Meta, or Merchant Center APIs, or ad-tech / e-commerce domain experience.
NestJS, tRPC, or GCP (Cloud Run, Artifact Registry, Workload Identity) in a previous role.
How we work
Small team, short feedback loops, no ceremony for ceremony's sake. Strict TypeScript, code review on everything, CI that deploys to Cloud Run on merge. We optimize for shipping value to customers weekly and keeping the codebase a place people enjoy working in.