Employment Type: 6-Month Contract / Internship
Location: Zurich, on-site (hybrid)

We are seeking a Junior Machine Learning Engineer / Intern to join our team, assisting in the post-training and optimization of large language models (LLMs), their integration with retrieval-augmented generation (RAG), and multi-agent architectures. The candidate will support improving our LLM-based solutions through applied research, solid implementation, evaluation, and optimizations. You will collaborate closely with experienced machine learning engineers, software engineers, and domain experts to push the frontier of LLM for code generation.

Key Responsibilities:

• Assist in designing LLM training datasets for instruction following, question answering, reasoning, conversational models, and code generation.
• Support the integration of curated content into fine-tuning pipelines.
• Help define instructions for knowledge workers to create LLM training data (code, text, prompts, dialogues) and synthetic data validation processes.
• Support systematic evaluation of LLM performance and benchmarking.
• Participate in refining retrieval-augmented generation (RAG) engines.
• Learn and assist in developing multi-agent architectures and prompt engineering.
• Experiment with new trends and research directions in LLMs under guidance.
• Implement LLM components and integrations in high-quality code, with proper documentation.

Required Qualifications:

• Currently pursuing or recently completed Bachelor’s or Master’s degree in Computer Science, Data Science, or a related field.
• Familiarity with machine learning frameworks (e.g., PyTorch or TensorFlow).
• Basic proficiency in Python.
• Strong communication skills and ability to collaborate effectively.
• Strong analytical and problem-solving abilities.

Preferred Qualifications:

• Basic understanding of LLM architectures like BERT, Llama-3, or similar.
• Familiarity with code generation or low-resource programming languages.
• Interest or experience in reinforcement learning from human feedback (RLHF) or reinforcement learning from verifiable rewards (RLVR).
• Familiarity with data annotation and labeling tools.
• Interest in bias and fairness issues in AI.
• Experience or interest in TypeScript.