We're hiring a Senior Fullstack AI Developer to build AI-powered learning products on Node.js and React. You'll own features end to end, from the LLM orchestration layer that runs tutoring, grading and content generation to the React screens students and teachers use every day.
Key responsibilities
Design and build AI orchestration pipelines that chain LLM calls, tools, retrieval and memory for learning use cases: AI tutors, auto-grading, feedback on essays, quiz and lesson generation, learner Q&A over course content.
Build multi-agent and workflow-based flows (planner, tutor, evaluator agents) with clear handoffs, retries, fallbacks and human-in-the-loop review for teachers.
Implement RAG over course materials, textbooks and LMS content: chunking, embeddings, vector search, re-ranking and citation back to source.
Develop Node.js (TypeScript) backend services and APIs that serve AI features at scale, including streaming responses, queues and background jobs.
Build React front ends for learners, teachers and admins: chat-based tutors, adaptive lesson views, dashboards and content authoring tools.
Set up evaluation, guardrails and monitoring for AI output: accuracy tests, hallucination checks, age-appropriate content filters, prompt versioning, cost and latency tracking.
Integrate with EdTech systems such as LMS platforms (Moodle, Canvas, Google Classroom), LTI 1.3, SCORM/xAPI and SSO.
Protect student data in line with FERPA, COPPA, GDPR and local data privacy rules.
Work with product, instructional designers and teachers to turn learning goals into features, and mentor mid-level engineers through code reviews and design discussions.
What you bring
Must-have requirements
7+ years of professional as Fullstack engineer in software development in the IT industry, with strong background in Backend side.
Strong EdTech experience: you've shipped products for schools, universities, online course platforms or corporate training, and you understand learners, teachers and admins as distinct users.
Hands-on AI orchestration in production (required). You've built LLM workflows or agents with tools such as LangChain / LangGraph, LlamaIndex, Vercel AI SDK, Mastra, CrewAI, AutoGen or a custom orchestrator, and can explain the design trade-offs you made.
Solid experience with LLM APIs (OpenAI, Anthropic Claude, Google Gemini or open-source models): prompt design, function / tool calling, structured output, streaming, token and cost control.
RAG experience with at least one vector store (pgvector, Pinecone, Weaviate, Qdrant, Elasticsearch or similar).
Advanced Node.js with TypeScript: Express, NestJS or Fastify; REST and/or GraphQL; async processing with queues (BullMQ, RabbitMQ, SQS).
Advanced React: hooks, state management (Redux Toolkit, Zustand or React Query), Next.js, responsive and accessible UI (WCAG).
Databases: PostgreSQL or MySQL, plus MongoDB or Redis.
Cloud and DevOps basics: AWS, GCP or Azure; Docker; CI/CD pipelines.
Strong system design experience: you've designed scalable, multi-tenant architectures (microservices or modular monolith, caching, queues, event-driven flows) and can whiteboard an AI-heavy platform end to end, including LLM latency, cost and failover.
Client-facing experience: you've worked directly with clients or stakeholders to gather requirements, present technical solutions and demos, estimate scope and handle feedback without a middle layer.
Good English, written and spoken, for daily communication with international clients and product teams.
Nice to have
Experience with adaptive learning, learning analytics or knowledge tracing (mastery models, spaced repetition, personalized learning paths).
Model Context Protocol (MCP) servers or agent-to-agent integrations.
Speech and multimodal AI for language learning or reading practice (Whisper, text-to-speech, image understanding).
Fine-tuning or running open-source models (Llama, Mistral, Qwen) on your own infrastructure.
Python for data and ML scripts alongside the Node.js stack.
Real-time features with WebSockets or WebRTC for live classes and collaboration.
Prior work as a tech lead or on a startup team that took an EdTech product from zero to launch.
