AI Full Stack Engineer at Mem0 in San Francisco, CA

Job description

About Mem0 We're building the memory layer for AI agents. Think long-term memory that enables AI to remember conversations, learn from interactions, and build context over time. We're already powering millions of AI interactions for enterprise customers and an open-source community of 160,000+ developers. We're a Y Combinator-backed company that has raised a $24M Seed + Series A from top-tier investors. This is a full-time, onsite role based in San Francisco, CA. You'll work closely with a highly technical team in person, where rapid collaboration and fast iteration are part of our culture. This position reports directly to the CTO, making it a highly visible role with significant influence on product direction, engineering execution, and the future of the Mem0 platform. Role Summary Build the AI product surface of the Mem0 Platform, the memory infrastructure powering the next generation of AI agents. You'll own the frontend end to end. We don't have a designer, so the person in this seat sets the visual and interaction bar for everything our users touch: the dashboard, the playground, the docs experience, the onboarding flow. You'll decide what it looks like, build it in Next.js and React, and wire it to the Python services behind it. Our users are developers building agents, so the product you ship is the thing they judge us on. You'll work directly with the CTO, research, and customers to turn real memory problems into interfaces that feel fast and obvious. This is a role for an engineer who has shipped LLM products to real users, has taste, and gets energy from owning the whole surface. What You'll Do • Own the frontend: Design and build the Mem0 dashboard, playground, and developer surfaces in Next.js and React. You set the design direction, build the component system, handle streaming responses and live state, and hold the quality bar. • Ship AI features end to end: Design APIs, build retrieval and memory pipelines, own the data models, and deploy to production. • Build with LLMs in production: Work with embeddings, vector search, RAG, function calling, and agent orchestration. Handle the messy parts: context windows, token budgets, non-deterministic output, graceful degradation. • Make quality measurable: Build evals, golden datasets, and regression suites for AI behavior so that model and prompt changes ship with evidence behind them. • Build for scale and speed: Optimize retrieval latency, caching, database queries, bundle size, and inference cost across the stack. • Own quality: Write tests, maintain strong typing and linting standards, review code thoughtfully, keep the codebase clean. • Collaborate deeply: Partner with Research to bring new memory capabilities into the product, and talk to customers directly to sharpen what you build. • Operate what you build: Implement observability for both application and model behavior, monitor production, troubleshoot, and keep improving reliability. • Lead with product sense: Make thoughtful technical and product tradeoffs, prioritize well, iterate on user feedback and data. • Go beyond your lane: Unblock teammates, pick up new technologies fast, do what it takes to deliver. Minimum Qualifications • 5+ years building and shipping production-grade full-stack applications at scale. • Deep frontend craft in Next.js, React, and TypeScript. You've owned the look and feel of a product in production, with examples you can show us. • Strong design sense and the ability to work without a designer. You can take a rough idea to a polished, accessible interface on your own, with a point of view on layout, typography, spacing, and motion. • Hands-on experience shipping AI-native products to real users using LLMs, RAG, embeddings, or vector databases. We want to hear about what you built, who used it, and what broke. • Strong proficiency in Python and modern web application architecture. • Experience with REST APIs, PostgreSQL, Redis, background job systems, and modern backend frameworks.