We’re looking for a motivated Junior Generative AI Developer to join our newly launched IT Pod project. This is a hands-on individual contributor role where you’ll collaborate with senior engineers to design, implement, and optimize cutting-edge Generative AI solutions. You’ll work with technologies like LLMs (GPT-4, Claude, Gemini), diffusion models, and multimodal systems — all while following ethical AI practices.1.Model Development & Fine-TuningAssist in training and fine-tuning generative models (text, image, code) using PyTorch, TensorFlow, or JAXImplement RAG (Retrieval-Augmented Generation) pipelines and optimize prompts for specific domains2. Tooling & IntegrationBuild applications using LangChain, LlamaIndex, Hugging Face TransformersIntegrate GenAI APIs (OpenAI, Anthropic, Mistral) into enterprise workflows3. Prompt EngineeringDesign and test advanced prompting strategies (few-shot, chain-of-thought, ReAct)Create reusable prompt templates for workflows like customer support, code generation, and content moderation4. Evaluation & OptimizationDevelop metrics for hallucination reduction, output consistency, and safety alignmentOptimize inference costs using quantization, distillation, or speculative decoding5. CollaborationWork with cross-functional teams (product, data, UX) to deploy AI solutionsDocument processes and contribute to knowledge-sharing sessionsQualificationsEducation:Bachelor’s or Master’s in Computer Science, Data Science, or related fieldTechnical Skills:Proficiency in Python and familiarity with PyTorch or TensorFlowBasic understanding of NLP and neural architectures (Transformers, GANs)Experience with cloud platforms (AWS SageMaker, GCP Vertex AI, Azure ML)Familiarity with prompt engineering tools (LangChain, DSPy, Guidance, LMQL)Experience with deployment tools (FastAPI, Docker, MLflow)AI/GenAI Exposure:Experience with at least two of the following:Hands-on projects with LLMs or diffusion modelsVector databases (Pinecone, Milvus) and orchestration toolsFine-tuning LLMs (Llama 2, Mistral) using LoRA, QLoRA, RLHFBuilding RAG pipelines with embedding models (BERT, OpenAI)Developing applications with Stable Diffusion, DALL·ENLP projects using spaCy or NLTKSoft Skills:Strong problem-solving mindset and curiosity about emerging AI trendsAbility to explain technical concepts to non-technical stakeholdersPreferred QualificationsCertifications:Microsoft Certified: Azure AI Engineer AssociateGoogle Cloud Professional Machine Learning EngineerWhat you can expect from us:Flexible work hours (we start between 7:00 and 10:30 am).8-hour work time with a lunch break already included - spend the rest of the day doing what is important to you as intended in the #2h4Family program.Hybrid work model (50/50)Integration events - monthly delicious breakfasts, movie nights, board game nights, outdoor events.Lively and modern office in the City Centre with parking space for employees.A supportive and friendly atmosphere created by passionate people.