We know what it takes to turn dreams of the digital future into products that people love. Finding, developing, and accelerating new digital businesses: we aim for the best technological solution to develop unique products and services for Porsche customers and beyond. We aim to find and scale new digital business models as well as optimize existing products.
We Porsche Digital, are a 100% subsidiary of Porsche. This means we are driven by courage and passion, we are committed to learning and improving, and we believe in teamwork and respect as the key to creating a great Porsche experience for the customer and our team.
- You plan, develop and implement large language model (LLM) and AI solutions to automate tasks such as text generation, image processing. Further, you implement data analysis, ML algorithms as well as associated workflows to replace repetitive tasks with machine support.
- Introduce the latest advancements in language models to our products, such as retrieval-augmented generation (RAG) techniques, and prompt engineering methodologies.
- Develop, fine-tune, and optimize language models using frameworks like TensorFlow, PyTorch, or Hugging Face Transformers leveraging one of the prominent cloud providers (AWS, MS Azure).
- Experiment with different pipelines, architectures, training algorithms, and hyperparameters to enhance model capabilities such as fluency, coherence, and relevance.
- Optimize models for inference speed, memory footprint, scalability, and safety (incl. ethical concerns).
- Design effective pipeline, prompts, and inputs to guide language model responses toward desired outcomes.
- Collaborate with domain experts to understand specific use cases and requirements to develop AI solutions.
- Evaluate the performance of different prompts using metrics such as response relevance, diversity, and user satisfaction.
- Deploy trained / fine-tuned language models into production environments, ensuring robustness, scalability, and reliability.
- Collaborate with teams of LLM & ML researchers in the development of full-stack, GPU-accelerated data preparation pipelines for multimodal models. Implement benchmarking, profiling, and optimization of innovative algorithms in Python in various system architectures, specifically targeting LLM applications.
- Monitor the performance of the implemented AI systems and continuously develop optimization strategies.
- Advanced degree in Computer Science, Computer Engineering, or a related field (or equivalent experience)
- Working experience in developing ML-based solutions from the exploration to the production phase
- Proven experience in the development and implementation of LLM, AI and process automation
- Experience with LLMs and RAG pipelines, MLOps
- Familiarity with ML (Machine Learning) technologies and frameworks such as Pandas, TensorFlow, PyTorch, Tensorboard, Jupyter, NumPy, SpaCy as well as experience in data analysis
- Ability to write well-structured and tested code in Python
- Firm grip on the statistical and mathematical concepts underlying data science methods
- Passionate problem solver, ability to think independently and proactively to find solutions
- Excellent knowledge of English, German is a plus
- Advanced problem-solving skills
- Experience from 1-2 projects implementing RAG and LLMs from PoC to production environments
- Expertise with at least one programming language additionally to Python
- Basic understanding of cloud technologies, and how to scale and run machine-learning solutions on the cloud (preferably AWS)
- Expertise with Git, GitLab, and CI/CD pipelines on GitLab
- Fundamental knowledge of Atlassian products like Jira and Confluence
- Knowledge in the field of Spark / Hadoop / Kafka / Hive / Impala / Hue
- Experience with Data Visualisation with Python or tools like Tableau or Qlik
- Knowledge in frameworks like MLflow, KubeFlow, and serving frameworks
- Experience with developing complex software which has been brought into productive use
- Experience with working in large teams using professional software development practices, such as writing clean code, participating in code review, using source code management, and using build and deployment pipelines
- Experience working in diverse, international teams
- Expertise in one application domain (preferably, but not limited to, automotive)
- Competitive payment
- Thirty d