Senior AI Backend Engineer

About your team You’ll join AI Engines – the team building the backend “brain” of AI Agents in Text App. These aren’t your average scripted bots. Our AI Agents understand context, communicate naturally, and take meaningful actions: from digging through the Knowledge Hub to answer customer queries, to guiding human agents, or triggering automations in third-party tools via Copilot. To make this happen, we need to design highly scalable backend systems capable of handling millions of conversations daily. We build the APIs, develop the reasoning layers, and continuously push the boundaries of what our agents can actually do. You will be responsible for: • Hands-on building, scaling, and maintaining the core backend architecture (Vector Database, AI Engines, Knowledge Hub, etc.). • Designing and implementing an Agentic reasoning and orchestration layer using Skills and tool calling (a core building block of our architecture). • Modernizing the core by taking full ownership of migrating legacy implementations into our modern, highly scalable AI ecosystem. • Collaborating with our AI researchers to experiment with, validate, and confidently push new AI features straight to production. • Mastering Context Engineering & R&D by moving way beyond basic prompt engineering. You'll be experimenting with and optimizing how our agents communicate with LLMs (taking full ownership of multi-turn conversations, tool calling, advanced RAG integrations, system prompt design, and history management). • Designing RESTful APIs and service integrations for AI and data-heavy use cases, while migrating legacy implementations to a modern ecosystem. • Integrating AI Agents with MCP servers, shaping how they interact with external ecosystems.  • Setting the standard by taking an active part in architecture discussions, code reviews, and technical decision-making, while bringing an extreme ownership and "fast flagging" mindset to the team. You may be a good fit for this role if you: • Have 6+ years of experience in backend development, with strong, recent, hands-on expertise in Node.js and TypeScript in recent projects. We need someone who is fluent in async programming, queues, and streams, and can jump into our codebase on Day 1.  • Have hands-on experience with real-time communication (WebSockets, async service-to-service messaging). • Have proven commercial experience building AI-driven features. You don't just know the concepts—you’ve actively implemented RAG, Function Calling, and Agentic systems in production.  • A deep understanding of Context Engineering. You know that it’s no longer just about "writing good prompts." You have proven experience managing context windows, multi-turn conversation history, and tool calling within Agentic or RAG ecosystems. • Have a solid grasp of database ecosystems, including SQL (Postgres), NoSQL (Elasticsearch), and Vector Databases (e.g., Pinecone, pgvector), along with the mechanics of embeddings. • Know how to run and scale services on Google Cloud Platform and Kubernetes.  • Enjoy solving complex technical challenges in microservices, concurrency, and performance optimization. Bonus if you also: • Experience with Go (Golang) is a strong plus. • Have experience building Data Pipelines or ETL processes. • Previous work on SaaS products, low-code/no-code platforms, or workflow automation. Our Tech Stack: (Let’s be transparent: while we value curiosity and a growth mindset, solid hands-on experience with Node.js and TypeScript is a strict requirement for this role. We are happy to give you time to learn our specific AI frameworks and databases on the go, as long as your core backend skills are top-notch.) • Core (Must-have): Node.js, TypeScript • Databases: PostgreSQL, Redis, ElasticSearch • Scalability & communication: PubSub, Webhooks, SSE • AI-related: LLMs, LangGraph, Function Calling, Context Compression, MCP (Model Context Protocol)  • Infrastructure & deployment: Docker, GitHub Actions, Prometheus, Grafana • Testing & reliability: Vitest, Supertest, Playwright How we work We’re a hybrid team – remote when it makes sense, onsite when it matters most. Once per quarter, your whole team spends 5 working days together at our Wrocław office – a time for focused execution, unblocking work, and driving key outcomes. Our office is open year-round – drop by anytime. Contract: B2B (self-employed). You’ll need a registered business in selected EU countries — either already set up or possible to register in your country. Please apply only if this works for you.