Your responsibilities:
· Prototyping ideas and evaluating how they would fit into our product vision.
· Maintaining a balance between cutting-edge research and practical applications, producing deliverables and products that set industry benchmarks.
· Stay updated on the latest advancements in RL, NLP and machine learning, ensuring our solutions remain at the forefront of technology.
· Model Development and Fine-tuning: Implement, refine, and fine-tune state-of-the-art model architectures, ensuring they perform in real-world scenarios. Design and implement RL algorithms to fine-tune LLMs, focusing on improving performance in real-world applications.
· Documentation and Reporting: Maintain detailed records of AI experiments, findings, and methodologies, communicating complex insights to varied audiences.
Your profile:
· You care about making something people want. You want to ship something that will bring value to our users. You want to deliver AI solutions end-to-end and not end on building a prototype.
· Degree in Computer Science or a related field.
· Demonstrated experience in developing and deploying RL algorithms, preferably in the context of natural language processing or LLMs (e.g. RL from human or AI feedback, LLM alignment, DPO, PPO, multi-agent systems).
· Familiarity with popular NLP tools and frameworks such as PyTorch or HF transformers. Prior experience with distributed training tools like Ray is a plus.
· In-depth knowledge of transformer architectures.
· Experience with research organizations and structured work.
Nice if you have:
· Experience with automation of prompt engineering semantic search and multi-modal models. Experience with human in the loop systems.
· Experience with agentic systems
· PhD in Computer Science or a related field.
· Publication track record.
What you can expect from us:
Be part of an AI revolution!
30 days of paid vacation
Access to a variety of fitness & wellness offerings via Wellhub
Mental health support through nilo.health
Substantially subsidized company pension plan for your future security
Subsidized Germany-wide transportation ticket
Budget for additional technical equipment
Flexible working hours and a hybrid working model for better work-life balance
Virtual Stock Option Plan
