TimotheBoulet

PhD student at Inria (Flowers team) on LLM for environment-agent interaction

[TimotheBoulet @TimotheBoulet]

Repo: https://t.co/D9LaK7Yn1t

Works with Anthropic direct, any LiteLLM compatible provider, and local models (Ollama / vLLM / SGLang).

Hope you will like it ! Feedback very welcome 🤖

[TimotheBoulet @TimotheBoulet]

I want to share Alan Code, my own implementation of a Claude Code like agent 🤖

What it adds:
- Designed as a Python library to build your systems on, not just a CLI
- Browser GUI with an LLM Perspective panel and Git integration
- Project and global memory + live cost tracking

[TimotheBoulet @TimotheBoulet]

The GUI mode opens a 3-panel browser view:

- Alan Chat : a cool chat interface, equivalent to CLI mode
- LLM Perspective : the model's exact view: full prompt, tool results, compaction, ... great for prompt debugging
- Git Tree : move or revert conversation along the git tree

[TimotheBoulet @TimotheBoulet]

I will be presenting my poster at the Efficient Reasoning Workshop on the 6th December at #NeurIPS2025 ! If you're interested in LLM agents on complex interactive code-environments, I would be happy to chat !

[TimotheBoulet @TimotheBoulet]

Excited to share our new work at #NeurIPS2025 !

📄Software Engineering Agents for Embodied Controller Generation

LLM-based SWE-Agents are great at classic coding tasks, but what happens when the program to write isn’t static code, but a controller that must act step-by-step ?

[TimotheBoulet @TimotheBoulet]

Controller generation pushes SWE-Agents into a new domain where they must reason sequentially, uncover environment structure, and adapt through interaction — going beyond traditional software-engineering benchmarks. Poster here :

[Clément Moulin-Frier @ClemMoulinFrier]

@MaxTaylorDavie1 will present our new paper at the @EvostarConf on Thu.

Emergent Kin Selection of Altruistic Feeding via Non-Episodic Neuroevolution
https://t.co/BUUslHeVrH

With @hamongautier and @TimotheBoulet
, leveraging previous work with @nisioti_eleni

Summary:

[Matt Shumer @mattshumer_]

Wow. This is huge.

The first time (I'm aware of) that an OpenAI exec has publicly stated that they believe OpenAI is clearly prioritizing capabilities over safety research.

Massive implications, in many ways.