Infinite Garden · New York · Remote

About Infinite Garden

Most companies are built like machines: rigid structures, fixed handoffs, information trapped between functions, and learning repeatedly lost.

Infinite Garden is building something different - companies that can learn what they should become. We believe companies should be able to sense their environment, respond, experiment, remember, and grow toward opportunity.

That doesn't mean one intelligence doing everything. It means societies of intelligences with different roles, histories, skills, and points of view, working with people to create and operate real consumer brands.

Infinite Garden creates and operates consumer brands with AI, end to end. Starting with a product concept, our systems research the market, develop the brand, produce packaging and creative, buy media, run storefronts, take orders, and coordinate fulfillment. The brands are real: they spend on Meta, post on TikTok, take payments through Stripe, and ship from a warehouse.

The platform learns from its own work. Quality verdicts and run outcomes become telemetry. Nightly agents turn that evidence into lessons, feed the strongest lessons back into the system, and measure whether each belief helped or hurt. Validated lessons become bets with real ad spend, explicit kill criteria, and calibrated forecasts. When an agent proposes changing its own instructions, the change is tested against production and opened as a pull request for a human to review.

That learning loop is beginning to run. You'll improve it, make it more reliable, and extend what it can safely do.

About the role

We’re looking for a Senior AI Software Engineer to own and expand entire areas of the product. You'll build, lead decision-making, know when to bring others in, and use AI to increase your efficiency. You'll also work closely with the team to put the systems, processes, and safeguards in place that keep speed from becoming chaos.

If you've built products from 0 to 1 and want to innovate on the cutting edge of where AI, creative, and commerce meet, this is a chance to be part of the future of brand-building.

What you'll own

• Ship across the full system. Take ambiguous product and technical problems from first principles to production across APIs, workers, data systems, and the frontend.
• Extend the learning loop. Improve how agents generate lessons, form bets, evaluate results, and turn evidence into safe changes.
• Make agent systems reliable. Design for observability, idempotency, permissions, recovery, and long-running orchestration where failures affect real customers and real money.
• Build roads for other builders. Create permission scopes, review paths, versioning, canary rollouts, and tools that let agents and non-engineers contribute safely.
• Set technical direction through shipped work. Make architecture calls, document them, bring others with you, and remove complexity where possible.

The stack: TypeScript end to end. Hono API and BullMQ workers on Railway. Postgres with Drizzle, ClickHouse, Redis, and GCS. Next.js and React 19. LangChain Deep Agents with Anthropic, OpenAI, and Google models. Vitest and Playwright. OpenTelemetry, Sentry, and LangSmith.

What success looks like

In your first 30 days, you will have: