Senior Full-Stack Engineer

About Understood Understood is a nonprofit focused on shaping the world for difference. We raise awareness of the challenges and strengths of people who learn and think differently. Our resources help people navigate challenges, gain confidence, and find support and community so they can thrive.  Having a shared commitment to our values is a key factor in any hire we make. We have five core values:  Continuously learn. Act with intention. Champion difference. Inspire change. Grow together.  Come be part of an organization with an entrepreneurial spirit that’s helping to shape the world for difference. Together, we can build a world where everyone can reach their full potential.  To learn more about Understood, please visit: www.understood.org . Who you are  To advance Understood’s mission and further our impact, we’re hiring a Senior Full-Stack Engineer, reporting to an Engineering Manager, to work across our frontend, backend, and AI systems. You’ll join a cross-functional team and collaborate closely with product, design, data, and content teams to build and evolve the platforms that power Understood.org and our AI-assisted experiences. The ideal candidate is comfortable working across the stack—from React and Next.js on the frontend to Python and Django on the backend—and brings demonstrated experience with LLM-powered products. You’ll contribute to a growing portfolio of AI tools, including our conversational assistant and emerging multimodal experiences, while helping millions of families navigate learning and thinking differences. This is a hybrid role that requires a minimum of three (3) days a week in our NYC office, with a focus on Monday, Tuesday, and Thursday. We are a Mac-based environment, and team members are provided a Mac laptop for their work.  What you’ll do Collaborate with cross-functional teams (product, design, data, content) to design, develop, and maintain features across the full stack Build high-quality, accessible frontend experiences in Next.js, React, and TypeScript, including the stateful, streaming chat and multimodal (voice, text-to-speech, speech-to-text) interfaces that power our AI assistants. Architect and ship the backend services and APIs (Python, Django, SQL) behind them, including integrations with PostgreSQL, Redis, Pinecone, and Contentful. Monitor and improve AI performance KPIs, including system latency and time-to-first-token (TTFT) You know how to build accessible and localized products Own your work across the entire software development lifecycle, from architecture and planning through development, deployment, and monitoring Ship early and often. Break big problems into small, releasable increments and get them in front of users using feature flags, A/B tests, and progressive rollouts rather than big-bang launches. Stay close to the people we serve. Dig into user feedback, support tickets, and product analytics to find the real problem, then ship the smallest change that solves it. Leverage tools such as DataDog, Sentry, and Langsmith for monitoring, alerting, and analytics to proactively identify and resolve issues; own your services’ health Collaborate effectively with both internal teams and external development partners to co-design solutions, troubleshoot integrations, and review code Participate in code reviews, mentor junior engineers, and contribute to continuously improving development processes and best practices Stay current with emerging technologies and industry trends, advocating for tools and practices that enhance our development process and product offerings Must-haves  7+ years of professional experience delivering production software, with meaningful depth in both frontend and backend development Strong expertise in TypeScript, React, and Next.js, including experience building stateful, streaming interfaces Demonstrated  experience with Python and Django (or equivalent backend frameworks) Proficiency in designing and building RESTful API’s Experience working with relational databases (PostgreSQL, MySQL) and vector databases (Pinecone) Comfortable working in the cloud (GCP, AWS, or similar) Experience monitoring and instrumenting production services; we use Datadog and expect engineers to own their services’ health Demonstrated ability leading projects as a technical lead Ability to break complex projects into small, shippable increments and get them in front of real users quickly, using feature flags, staged rollouts, and experimentation platforms (Statsig or similar) to de-risk along the way, rather than sitting on a polished v3. Genuine interest in user problems. You read the support inbox, watch session replays, and treat "is this actually working for people?" as part of the definition of done. Excellent communication skills, particularly when explaining technical matters to non-technical stakeholders and more junior engineers Thrives in a collaborative environment and is motivated to promote a culture of teamwork, knowledge sharing, and continuous improvement Familiarity with observability tooling beyond basic monitoring (distributed tracing, log analytics) Nice-to-haves AI & Personalization Production experience with LLM orchestration and agent design: LangChain / LangGraph / LangSmith or equivalents, multi-agent routing, memory and personalization layers (LangMem or similar), and Model Context Protocol (MCP). A practical approach to making LLMs reliable in production: prompt engineering and tuning, systematic evaluation of outputs, guardrails and fallbacks, and designing around hallucinations and other failure modes. Experience with retrieval-augmented generation (RAG) and vector databases (Pinecone, Weaviate, pgvector, or similar). Experience building real-time, multimodal AI experiences (streaming text, voice, TTS/STT, Web Audio, WebRTC, real-time OpenAI SDKs), including optimizing latency and time-to-first-token (TTFT)