Tobias Lindenbauer is an AI researcher at JetBrains Research, where he advances efficient and effective code agents that robustly solve long-horizon software engineering tasks. His work sits at the intersection of large language models and developer tooling, asking how autonomous agents can plan, act, and manage their own context over the long stretches of work that real software engineering demands.

Currently, he is most interested in efficiency topics, context management, interpretability, and data synthesis. He recently presented The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management at the Deep Learning for Code workshop at NeurIPS 2025. The paper highlights practical pitfalls of LLM summarization-based context strategies and provides evidence for simpler, more computationally efficient alternatives, a contribution with direct implications for the cost and reliability of coding agents.

Through this research, Tobias helps push code agents from impressive demos toward dependable tools, combining empirical rigor with an engineer’s attention to what actually works in practice.