Senior AI Fullstack Engineer (Next.js & LLM)

About the Role At Blazity, we are a team of React.js and Next.js experts building modern, API-first digital products for enterprise clients. We are expanding our AI capabilities—autonomous agents, AI-driven automation, and LLM-powered workflows—and are looking for a Senior AI Fullstack Engineer who can design and ship production-ready AI systems. You will work at the intersection of modern web architecture and generative AI , taking ownership of solutions delivered to real users. Our requirements Must-have Technical Excellence: 8+ years of experience as a consultant / forward deployed engineer Strong experience with Next.js (App Router), React, TypeScript Deep understanding of: Server vs Client Components (RSC) streaming, caching, revalidation, Server Actions Vercel AI SDK / LangGraph / MCP Experience building RAG systems Claude extended thinking / prompt caching / Batch APIs AI Engineering: Proven experience shipping LLM-powered features to production Hands-on with Anthropic / OpenAI APIs (messages, tool use, streaming) Practical experience with: prompt engineering (structured output, few-shot, XML) tool calling / agent workflows multi-turn orchestration Understanding of: token economics (input/output costs, context window) latency vs quality trade-offs Reliability & Quality: Experience with: prompt evaluations (LLM-as-judge, rule-based) handling hallucinations and failure modes Ability to design robust AI systems , not just prototypes Soft Skills: Strong ownership mindset Experience working directly with clients Clear communication and structured thinking (English C1) Ability to make decisions in ambiguous environments Nice-to-have: What You Will Do Ship Production Apps: Build scalable web applications using Next.js (App Router) and TypeScript , ensuring high performance and clean architecture. Engineer AI Workflows: Develop and orchestrate LLM-powered features using Anthropic and OpenAI APIs, focusing on tool use and streaming. Agentic Orchestration: Design autonomous agent loops and multi-turn workflows, knowing when to use a structured pipeline versus a dynamic agent. Optimize & Evaluate: Manage token economics (pricing, context windows) and implement prompt evaluations (LLM-as-judge, rule-based) to ensure system reliability. Collaborate: Work directly with senior engineers, founders, and client teams to shape AI capabilities from concept to production. What we offer B2B: 120 – 180 PLN/h + VAT 🔥 100% remote (Poland) 29 paid vacation days Private healthcare + sports card High-impact work on real AI systems