AI Engineer

O projekcie: Hi there! 👋🏻 We’re Tooploox 💎, a Solvd Inc. company. We create AI-powered products and services that make a real difference. Our team of nearly 200 specialists - including a 40+ person R&D team with many PhDs - has pioneered AI solutions across industries from healthcare to e-commerce. We’ve published research in top-tier venues like NeurIPS and ICML, while delivering real-world applications that help our clients innovate and scale. We are seeking a dynamic and motivated AI Engineer with a strong focus on developing Production-Grade AI Agents. If you have a deep understanding of modern AI stacks and thrive on moving beyond simple demos to build reliable, evaluation-driven solutions, we want to hear from you. This role involves not only technical expertise but also collaboration with clients to understand and refine project requirements. How we work: At Tooploox 💎, you have the flexibility to choose your working hours ⏰ and location 📍. While we value remote work, we also believe in building relationships 🤝and invite you to join us in our Warsaw and Wrocław offices 🏢. Enjoy a relaxed atmosphere 🍃 and try some “home-made” pizza 🍕 from our office pizza oven. We love having pets 🐶 in the office, so feel free to bring yours along.😁 Join us and shape the future of AI while working the way you like! Wymagania: Experience and skills you need to join us: - Solid full-stack expertise in Python, with a focus on writing clean, modular code, complemented by experience with TypeScript or Node.js. - Deep practical experience with modern orchestration and state-management frameworks like LangGraph or LlamaIndex Workflows. - Strong understanding of Retrieval-Augmented Generation (RAG) including advanced techniques like hybrid search, re-ranking, and graph-based retrieval. - Experience implementing tracing and monitoring for complex agent flows using tools like LangSmith, LangFuse, or Arize Phoenix. - Proven ability to design Evals and test pipelines to prevent regression and hallucinations in production apps. - Experience with API/SDKs of major model providers (OpenAI, Anthropic, Gemini) as well as Open Source models. - Experience configuring services on cloud platforms (e.g., AWS, Azure, GCP) and containerization (Docker/K8s). - Proficient in automating CI/CD processes and understanding DevOps practices. - Openness to client collaboration, fostering clear communication and effective partnership. - Fluency in English (you will attend meetings with English speaking clients). It would be great if you also have: - Experience running and serving local LLMs (e.g., via Ollama, vLLM, or TGI). - Familiarity with programmatic prompting optimization tools like DSPy. - Hands-on experience with the Model Context Protocol (MCP). - Proficiency in Vector Stores (Pinecone, Qdrant, Weaviate, or pgvector). - Familiarity with Voice Agents concepts and frameworks like Pipecat. - Experience with graph databases (e.g., Neo4j) or GraphRAG approaches. - Fundamental knowledge of machine learning principles or model fine-tuning. Codzienne zadania: - Collaborate with cross-functional teams to design and build stateful, multi-agent workflows using modern orchestration patterns. - Engage with clients to define requirements and oversee the full software development lifecycle, from proof of concept through evaluation and implementation to production. - Implement rigorous evaluation pipelines (e.g., LLM-as-a-Judge) to quantitatively measure agent performance and reliability before deployment. - Proactively drive productivity and adapt to evolving demands, taking initiative consistently. - Manage backend tasks while contributing to frontend development to ship functional end-to-end prototypes. - Stay informed about rapid advancements in the field (such as reasoning models, SLMs, or new prompting paradigms), continuously learning and integrating new knowledge. - Employ a pragmatic, evaluation-driven approach to adopting innovative AI solutions.