AI / LLM Software Engineer
Blau Direkt Poland
⚲ Kraków
20 000 - 27 000 PLN (B2B)
Wymagania
- AI
- Clean Architecture
- DDD
- Clean Code
- Docker
- Git
- Communication skills
- Golang
- Elasticsearch (nice to have)
- Testing (nice to have)
- MySQL (nice to have)
- MariaDB (nice to have)
- React (nice to have)
- TypeScript (nice to have)
- System integration (nice to have)
- JavaScript (nice to have)
- Unix (nice to have)
- GitLab CI (nice to have)
- Jira (nice to have)
- Confluence (nice to have)
- Slack (nice to have)
- Sentry (nice to have)
Opis stanowiska
O projekcie: AI / LLM Software Engineer - backend developer About the Role We are looking for an experienced AI / LLM Software Engineer with a strong backend background in Goland to help us build production-ready AI capabilities. This is not a pure “Prompt Engineer” role. We are looking for someone who understands how to design and build reliable systems around LLMs: RAG pipelines, AI agents, model integrations, observability, cost control, and AI-specific security. Go will be one of the main implementation languages, but the core focus is on building robust, scalable, and secure AI-powered systems. Our Tech Stack- Languages: Go, TypeScript, JavaScript, PHP- AI / LLM: OpenAI, Anthropic / Claude, Gemini, Cursor, JetBrains AI, RAG, vector databases, embeddings, hybrid search, AI agents- Databases: MySQL, MariaDB- Infrastructure: UNIX, Kubernetes, RabbitMQ, Docker- CI/CD: GitLab CI, CodeMagic- Testing: Playwright, Python, Dart, Postman- Tools: Jira, Confluence, Slack, G-Suite, Sentry Benefits and Perks - Private healthcare.- Cafeteria system.- Training budget: €800 per year.- Internal trainings, hackathons, integration trips, and team meetings.- Referral bonus.- Startup atmosphere.- Flexible working hours.- Remote or hybrid work.- Modern office with chillout zone, standing desks, bicycle parking, cloakroom, and showers.- Regular office lunches and integration meetings.- Some international travel included. Wymagania: Your Profile - Several years of experience in software engineering, ideally with a strong backend focus. - Practical experience with LLM-based applications, AI workflows, or AI-assisted products. - Good understanding of RAG concepts: vector databases, embedding models, chunking, retrieval quality, and hybrid search. - Experience integrating LLM APIs, working with prompts, structured outputs, and tool/function calling. - Awareness of AI-specific security risks, including prompt injection, PII stripping, data leakage, and unsafe tool execution. - Strong backend development experience, preferably with Go. - Solid understanding of software architecture, clean code, and system design. - Experience with Docker, Git, CI/CD, and production environments. - Ability to use modern AI development tools such as Cursor, Claude, Gemini, or JetBrains AI effectively. - Strong ownership, pragmatic mindset, and good communication skills. - Fluent English, minimum B2. Nice to Have - Experience with vector databases such as Qdrant, Pinecone, Weaviate, Milvus, Elasticsearch, or OpenSearch. - Experience with AI evaluation, retrieval quality testing, or automated testing of non-deterministic systems. - Experience with event-driven architectures and asynchronous processing. - Experience with Kubernetes, RabbitMQ, MySQL, or MariaDB. - Experience with React and TypeScript. - Knowledge of PHP or legacy system integration. - Knowledge of German. Codzienne zadania: - Design and build production-ready AI / LLM-based features and platform components. - Integrate LLM APIs and AI models such as OpenAI, Anthropic, Gemini, or local models. - Build and optimize RAG pipelines, including vector databases, embeddings, chunking strategies, and hybrid search. - Design and implement AI agents and agentic workflows using LLMs, tools, APIs, and internal data sources. - Develop backend services and APIs, primarily in Go and occasionally in PHP. - Monitor and optimize LLM costs, token usage, latency, reliability, and output quality. - Implement safeguards against prompt injection, data leakage, unsafe tool usage, and improper PII handling. - Design maintainable systems using clean architecture, DDD, and pragmatic engineering principles. - Evaluate new AI technologies and apply them where they provide real business value. - Collaborate with product and engineering teams to bring AI concepts from prototype to production.