AI Engineer (Langchain, PyTorch, AzureML)

Designing and operating AI-driven pipelines that automatically ingest, analyze, and classify operational incidents and tickets. The system leverages LLMs with structured prompt logic and few-shot prompting (using historically similar tickets) to deliver consistent, scalable triage — replacing manual classification and accelerating resolution. 

Codzienne zadania

• As an AI Engineer at DS Stream, you will bridge the gap between traditional data science and modern AI applications.
• Leveraging your strong foundation in machine learning, statistics, and data engineering, you will design, train, and deploy ML models while integrating GenAI capabilities into existing data pipelines.
• You will be responsible for the full ML lifecycle - from experimentation and model development to production deployment and monitoring - ensuring our AI solutions are robust, scalable, and aligned with business objectives.

Nice to Have:

• Familiarity with NLP techniques and transformer architectures
• Experience with LLM fine-tuning and adaptation techniques
• Knowledge of RAG systems and embedding models
• Familiarity with MLOps practices (model versioning, CI/CD, monitoring)
• Working knowledge of data pipeline tools (Apache Spark, Airflow, or similar)
• Experience with experiment tracking tools (MLflow, Neptune, Weights & Biases) Docker, Kubernetes, and containerized ML deployments
• Experience with A/B testing and model performance optimization

Must have

• AI
• Data science
• Machine Learning
• Data pipelines
• Python
• NumPy
• pandas
• PySpark
• scikit-learn
• Deep learning
• PyTorch
• TensorFlow
• SQL