The Heuristic Scientist: Open-Ended Algorithm Discovery with LLMs

​We are excited to host a BLISS x juna.ai workshop featuring Dr. Nima Siboni (senior RL Engineer), who will guide us through an interactive session on algorithm discovery with LLMs.

​​​ Title: The Heuristic Scientist: Open-Ended Algorithm Discovery with LLMs
📅 Date: 03.06.2026
🕕 Time: 18:00
📍 Location: TU Berlin Marchstrasse 23 [Room 0.011]

​​​The session will last around 2 hours, followed by a networking session with juna.ai and fellow AI enthusiasts (and free pizza!🍕). Bring your laptop!

​Abstract:Large language models are increasingly useful not only for writing code, but for exploring the space of possible algorithms. In this hands-on workshop, we will look at how LLMs can act as “heuristic scientists”: generating candidate strategies, turning ideas into executable code, testing them against concrete objectives, and refining them through feedback. The workshop centers on the general discovery loop behind LLM-assisted heuristic generation. Participants will learn how to structure problems for algorithm discovery, prompt models to propose interpretable heuristics, evaluate those heuristics with simulations or benchmarks, and use the results to guide further search. We will also discuss when simple prompting is enough, when training or fine-tuning may help, and how to keep generated heuristics understandable, testable, and auditable.

​By the end of the session, participants will have built a small discovery loop of their own and will understand the practical trade-offs involved in using LLMs for open-ended algorithm discovery: creativity versus control, exploration versus verification, and automation versus human judgment.

​We are BLISS e.V., Berlin’s AI community connecting like-minded individuals passionate about machine learning and data science. Our BLISS Workshops connect students and young professionals with industry partners, offering an inside look into how machine learning is applied in real-world settings - from research and development to deployment.

​

​​​BLISS Website: https://bliss.berlin

​​​BLISS Youtube: https://www.youtube.com/@bliss.ev.berlin