Federico Cassano (@ellev3n11) / X

Pinned

Oct 29, 2025

so excited to share Composer with the world! Composer is Cursor's own agentic coding model. it plans, edits, and builds software alongside you with precision, keeping you in flow with incredible speed. i started this project on the side while working on a bug-finder prototype

This post is unavailable.

Federico Cassano

@ellev3n11

Sep 9, 2025

incredibly grateful. thank you to everyone that helped me get here. especially to @cursor_ai for truly making it happen and to @ArjunGuha for being such a great inspiration.

Federico Cassano

@ellev3n11

Apr 29, 2024

huggingface.co/blog/sc2-instr… We finally released StarCoder2 Instruct! SC2-Instruct is the very first entirely self-aligned code LLM trained with a fully permissive and transparent pipeline. On benchmarks, we are beating even versions of StarCoder2 trained on GPT-4 distilled data!

StarCoder2-Instruct: Fully Transparent and Permissive Self-Alignment for Code Generation

Happy to share that I'll be taking a break from school to work at @cursor_ai as a research scientist! I'm extremely excited to continue my research on using AI and PL for automating software engineering with such a talent-dense team.

Federico Cassano

@ellev3n11

Oct 21, 2025

we got flash attention 4 varlen backward before gta 6

Federico Cassano

@ellev3n11

Jun 3, 2025

my rl codebase is multi-step, multi-objective, multi-reward, multi-environment, btw

CURSOR-BIG

Does anyone know someone who loves Slurm, lots of GPUs, and despises Kubernetes? Would love to chat with them

Federico Cassano

@ellev3n11

Sep 1, 2025

checked-out the puppies at @VoltagePark IRL. H100s are pretty blocky, unlike gaming graphics cards

Federico Cassano

@ellev3n11

Jun 4, 2025

BugBot has saved me from so many bugs that missed human review, people should try it! I found it to be especially useful for figuring out buggy edge cases in complex logic.

This post is unavailable.

Federico Cassano

@ellev3n11

Nov 5, 2025

Zichen Liu

@zzlccc

Oct 31, 2025

BF16 -> FP16 is such a simple (one configuration change in Oat) yet fundamental fix for inference-training mismatch. With FP16, the most basic importance sampling PG outperforms all algorithmic fixes in BF16. Let's rethink RL stability from the precision perspective.🔎

Federico Cassano

@ellev3n11

Oct 29, 2025

More on our blog post: cursor.com/blog/composer

Federico Cassano

@ellev3n11

Oct 29, 2025

so excited to share Composer with the world! Composer is Cursor's own agentic coding model. it plans, edits, and builds software alongside you with precision, keeping you in flow with incredible speed. i started this project on the side while working on a bug-finder prototype

Federico Cassano

@ellev3n11

Nov 1, 2025

2026/2027 will be the age of fault tolerance, health checks, and CP+DP->EP

Stas Bekman

@StasBekman

Oct 31, 2025

Remember how we were stuck with 80GB HBM for a really long time? This pattern is breaking - copious GPU memory will be the new norm in high end gpus in 26/27 - Nvidia Ultra Rubin - 1024GB HBM - Qualcomm MI200 - 768GB LPDDR - AMD MI400x - 432GB HBM So ML performance

Federico Cassano

@ellev3n11

Jul 25, 2024

Replying to @Calcolis and @cremieuxrecueil

Because we receive environmental feedback, in this Nature's paper, the model doesn't.