Job description

30tril is building an AI food assistant that works for the household, not the seller: an iPhone app that plans meals, builds the grocery basket and helps you shop, with no ads. They are looking to hire their first Founding Engineer. Location: New York preferred; remote considered. Type: Six-month contract converting to full-time at our first round. Founding team, equity-first. The role You'll take a working product from internal testing to a live beta with real households, then to scale. Today that product is an iPhone app with chat and voice, a Python/Django back end, and AI from several model providers behind one routing layer. You'll refine the architecture, set the technical roadmap with the founder, build the missing pieces and own problems end to end. You'll also shape product decisions, sit in on household research, and set the engineering standards future hires inherit. This is a product you and everyone in your life will use every day. Problems you'll work on There's no textbook for most of this. The work is hypothesis, test, measure and try again, with real families on the other end. • Trust: a recommendation engine a household can rely on for every food decision. • Catalog: turning messy, contradictory product data from hundreds of sources into one catalogue we can stand behind. • Matching: recognizing the same product across retailers, pack sizes and label changes, and keeping price and availability fresh across thousands of stores. • Household personalization: planning for several people with different needs, preferences and budgets. • Recipe to basket: turning any recipe link into a week's plan, a basket and step-by-step help at the stove. • Discovery: keeping it personal and current across creators, experts and recipes. • Agents: an assistant that acts for the household reliably enough that they stop double-checking it. • Quality: knowing an answer is right before a user sees it, and keeping it right as models and data change. • Cost and latency: balancing speed, cost and quality across model providers in real-time chat and voice. What you'll own • The whole stack: the problems above and everything beneath them, from data pipelines to the app. • Running it: building, deploying and monitoring everything yourself, with tests and quality baselines gating every release. • Shipping: App Store releases, reliability, and the safety and security standards real users depend on. • Creator tooling: end-to-end tools for creators to publish and operate on the platform. • Internal tooling: the systems the company needs as it builds out each function. Who we're looking for A hands-on full-stack engineer, strongest on the back end, who has shipped a consumer product and wants to own a company's engineering from the first line. What you bring • Back end and production: deep, recent hands-on work in Python (Django ideally) and Postgres. You read and change cloud infrastructure, run your own deploys, and have stories about the incidents. • Mobile: TypeScript and React Native with Expo. You can own a large iPhone app end to end and work in its thin Swift layer when needed. • AI in production: language-model features real users rely on, including tool calling, streaming, guardrails, and evaluations that gate every release, with an eye on cost and latency. • Shipped to real users: a consumer product people used, ideally an iOS app through App Store review. • Leadership: you've designed systems other engineers built on, and you've led engineers or are ready to. • Fluency with coding agents: you direct, review and debug their work and own both the architecture and the shipped result. How you work • Product judgment: the why leads the what and the how. You weigh user needs, incentives and commercial goals across experience, trust and business value. • Ownership: you make considered calls with incomplete information, use resources wisely, raise problems early and bring solutions. • Candor and respect: reasoned views, productive disagreement, owning mistakes and correcting course quickly. Nice to have • Product-catalog, entity-matching or data-ingestion work across many sources • Search or recommendations, including embeddings and vector search • Real-time voice • Food, grocery, nutrition, health or D2C commerce • A previous founding-engineer or first-engineer seat How we work We're AI-native across the product and the company. We fund the tools and models you need and give you plenty of autonomy to think, experiment and ship, with your code in front of real users from the start. We're remote-first, and you own your time and space. In person, we run on good food, walk-and-talks, meme breaks and the occasional robot massage. Your unnecessarily strong opinions on lunch, and everything else, are welcome.

Requirements

Must-haves

Type: Six-month contract converting to full-time at our first round. Founding team, equity-first.
This is a product you and everyone in your life will use every day.
Problems you'll work on
There's no textbook for most of this. The work is hypothesis, test, measure and try again, with real families on the other end.
Trust: a recommendation engine a household can rely on for every food decision.
Catalog: turning messy, contradictory product data from hundreds of sources into one catalogue we can stand behind.
Matching: recognizing the same product across retailers, pack sizes and label changes, and keeping price and availability fresh across thousands of stores.
Household personalization: planning for several people with different needs, preferences and budgets.
Recipe to basket: turning any recipe link into a week's plan, a basket and step-by-step help at the stove.
Discovery: keeping it personal and current across creators, experts and recipes.
Agents: an assistant that acts for the household reliably enough that they stop double-checking it.
Quality: knowing an answer is right before a user sees it, and keeping it right as models and data change.
Cost and latency: balancing speed, cost and quality across model providers in real-time chat and voice.
What you'll own
The whole stack: the problems above and everything beneath them, from data pipelines to the app.
Running it: building, deploying and monitoring everything yourself, with tests and quality baselines gating every release.
Shipping: App Store releases, reliability, and the safety and security standards real users depend on.

Nice-to-haves

Location: New York preferred; remote considered.
Product-catalog, entity-matching or data-ingestion work across many sources
Search or recommendations, including embeddings and vector search
Real-time voice
Food, grocery, nutrition, health or D2C commerce
A previous founding-engineer or first-engineer seat
We're AI-native across the product and the company. We fund the tools and models you need and give you plenty of autonomy to think, experiment and ship, with your code in front of real users from the start.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aipythonpostgrestypescriptreactswift

Platform: dover
Effort: medium