Adithya S K's banner
Adithya S K's profile picture

Adithya S K

@adithya_s_k • 16,646 subscribers

Scaling RL Envs @huggingface 🤗 • Founded @cognitivelab_ai Prev : Research @MSFTResearch • ML @apple • 22

Shorts

This is hugeeee for open-source RL environments. You can now discover environments across OpenEnv, Verifiers, Harbor, NeMo Gym, and more, all on the Hugging Face Hub. A common home for RL environments is long overdue. Excited for what’s coming next !

This is hugeeee for open-source RL environments. You can now discover environments across OpenEnv, Verifiers, Harbor, NeMo Gym, and more, all on the Hugging Face Hub. A common home for RL environments is long overdue. Excited for what’s coming next !

16,350 views

You can now do RL weight sync over HF Buckets. > Async GRPO + LoRA in TRL > train on one Job, serve vLLM on others > sync just the small delta adapter through a Bucket. Trainer, inference and storage, all on Hugging Face infra. Beautifully optimised loop by @DirhoussssiAmine 👇 3h27 → 53 min for the same reward.

You can now do RL weight sync over HF Buckets. > Async GRPO + LoRA in TRL > train on one Job, serve vLLM on others > sync just the small delta adapter through a Bucket. Trainer, inference and storage, all on Hugging Face infra. Beautifully optimised loop by @DirhoussssiAmine 👇 3h27 → 53 min for the same reward.

18,551 views

Introducing RL Environment Creator Skill Now any one can create RL environments $ npx skills add adithya-s-k/RL_Envs_101 > You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ... > the repo has live working examples of environments that your coding agent can reference > The skill is design to first understand what type of model you are training and create an environment while keeping that in mind ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks. Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.

Introducing RL Environment Creator Skill Now any one can create RL environments $ npx skills add adithya-s-k/RL_Envs_101 > You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ... > the repo has live working examples of environments that your coding agent can reference > The skill is design to first understand what type of model you are training and create an environment while keeping that in mind ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks. Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.

47,002 views

Videos

LIVE
1.2k

Anya Rossi

sweetdream.ai

SweetDream.aiSponsored•Livecam
Streaming Now

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

HD live stream
Exclusive private shows
1.2k viewers online

Free preview available • Premium content

We collaborated with @scaleAI to build a high-fidelity port simulation RL environment where AI agents tackle real-world logistics challenges
1:00
adithya_s_k's profile picture

We collaborated with @scaleAI to build a high-fidelity port simulation RL environment where AI agents tackle real-world logistics challenges

Adithya S K

61,306 views • 11 days ago

You can literally RL a 4B VLM to ace GeoGuesser 🌍 > Open-source code, RL environment, dataset, training setup, evals, and everything you need to reproduce it end-to-end > dropping soon!!
0:34
adithya_s_k's profile picture

You can literally RL a 4B VLM to ace GeoGuesser 🌍 > Open-source code, RL environment, dataset, training setup, evals, and everything you need to reproduce it end-to-end > dropping soon!!

Adithya S K

117,964 views • 1 month ago

Releasing SmolDataEnvs 🤗 5K+ Verifiable RL Environment tasks for hill-climbing small models in code and data science. Completely open source: environments, evals, training
0:29
adithya_s_k's profile picture

Releasing SmolDataEnvs 🤗 5K+ Verifiable RL Environment tasks for hill-climbing small models in code and data science. Completely open source: environments, evals, training

Adithya S K

55,392 views • 1 month ago

we RL trained a 4B VLM to play GeoGuesser > environment, dataset, training recipe, evals, and code are all open-source. > runs on a single A100 > beat gpt 5.4 mini, haiku, qwen 3.5 122B and came close to sonnet on evals
0:20
adithya_s_k's profile picture

we RL trained a 4B VLM to play GeoGuesser > environment, dataset, training recipe, evals, and code are all open-source. > runs on a single A100 > beat gpt 5.4 mini, haiku, qwen 3.5 122B and came close to sonnet on evals

Adithya S K

68,799 views • 1 month ago

Introducing RL Environment Explorer 🧭 There are now 12M+ RL task entries across RL environments on the Hugging Face Hub. You can now explore, visualize, and run RL environments across: OpenEnv • Harbor • MiMo • Verifiers • NeMo Gym All in one place
0:29
adithya_s_k's profile picture

Introducing RL Environment Explorer 🧭 There are now 12M+ RL task entries across RL environments on the Hugging Face Hub. You can now explore, visualize, and run RL environments across: OpenEnv • Harbor • MiMo • Verifiers • NeMo Gym All in one place

Adithya S K

27,593 views • 17 days ago

Now you can also create animations like Grant Sanderson 👉🏻 excited to introduce manim_skills $ npx skills add adithya-s-k/manim_skill This animation was created just by prompting 👇
0:52
adithya_s_k's profile picture

Now you can also create animations like Grant Sanderson 👉🏻 excited to introduce manim_skills $ npx skills add adithya-s-k/manim_skill This animation was created just by prompting 👇

Adithya S K

278,515 views • 9 months ago

Multi-harness RL training is coming to OpenEnv > pick a model. pick a harness. pick a sandbox. train > Claude Code. Codex. Gemini CLI. OpenCode. Pi. Kimi. OpenHands. and many more. > Async RL loop. fully open-source. end-to-end, harbor compatible dropping soon 👀
0:22
adithya_s_k's profile picture

Multi-harness RL training is coming to OpenEnv > pick a model. pick a harness. pick a sandbox. train > Claude Code. Codex. Gemini CLI. OpenCode. Pi. Kimi. OpenHands. and many more. > Async RL loop. fully open-source. end-to-end, harbor compatible dropping soon 👀

Adithya S K

26,631 views • 1 month ago

Excited to release TaskSmith 🔨 > a specialised harness for generating RL environments from code Not a general coding agent with a long prompt, but a specialised orchestrator. Every stage is built for one job: "turn a PR into the best RL environment possible"
1:00
adithya_s_k's profile picture

Excited to release TaskSmith 🔨 > a specialised harness for generating RL environments from code Not a general coding agent with a long prompt, but a specialised orchestrator. Every stage is built for one job: "turn a PR into the best RL environment possible"

Adithya S K

17,132 views • 24 days ago

7K+ RL environments from Xiaomi across coding, cyber, general, music, and more. You can now explore, visualize, and actually run them. Pick an environment → pick a model → run the rollout → watch exactly what happens RL environments are a lot more interesting when you can actually play with them.
0:58
adithya_s_k's profile picture

7K+ RL environments from Xiaomi across coding, cyber, general, music, and more. You can now explore, visualize, and actually run them. Pick an environment → pick a model → run the rollout → watch exactly what happens RL environments are a lot more interesting when you can actually play with them.

Adithya S K

17,971 views • 26 days ago

Introducing Vibemotion The first ever AI that turns a single prompt into stunning motion graphics and videos in minutes Vibemotion
1:54
adithya_s_k's profile picture

Introducing Vibemotion The first ever AI that turns a single prompt into stunning motion graphics and videos in minutes Vibemotion

Adithya S K

207,206 views • 1 year ago

Introducing Repo2RLEnv Turn any repository into runnable, verifiable coding environments built from real PRs and commits for coding-agent evaluation or RL training > uv pip install repo2rlenv
0:29
adithya_s_k's profile picture

Introducing Repo2RLEnv Turn any repository into runnable, verifiable coding environments built from real PRs and commits for coding-agent evaluation or RL training > uv pip install repo2rlenv

Adithya S K

67,436 views • 4 months ago

You can now turn PRs into RL environments at scale !!
1:01
adithya_s_k's profile picture

You can now turn PRs into RL environments at scale !!

Adithya S K

13,774 views • 1 month ago

Vision-based(Colapli) RAG is becoming popular, so we built a platform to compare: - Simple OCR RAG - VisionRAG - Colpali - Hybrid Colpali 🚀 Introducing VARAG – the Vision-First RAG Engine (Vision Augmented Retrieval and Generation).
2:19
adithya_s_k's profile picture

Vision-based(Colapli) RAG is becoming popular, so we built a platform to compare: - Simple OCR RAG - VisionRAG - Colpali - Hybrid Colpali 🚀 Introducing VARAG – the Vision-First RAG Engine (Vision Augmented Retrieval and Generation).

Adithya S K

96,933 views • 2 years ago

I read a lot of open-source code and that’s mostly how I learn to write better code. But it often takes a while to understand different projects their structure, how things are connected and GitHub’s UX isn’t that great for that. built "Gitvizz" to solve that exact problem it instantly turn your codebase into interactive graph so you can see how everything’s wired Check it out at - gitvizz. com ps: thats a real horse
0:59
adithya_s_k's profile picture

I read a lot of open-source code and that’s mostly how I learn to write better code. But it often takes a while to understand different projects their structure, how things are connected and GitHub’s UX isn’t that great for that. built "Gitvizz" to solve that exact problem it instantly turn your codebase into interactive graph so you can see how everything’s wired Check it out at - gitvizz. com ps: thats a real horse

Adithya S K

35,313 views • 1 year ago

Introducing Omnidocs - Unified Visual Document processing toolkit ✅ 6+ document processing tasks ✅ 15+ Model supported ✅ Runs on consumer GPU/ Mac ✅ Supports multiple backend - Pytorch, vllm, mlx, open responses API Get started → pip install omnidocs✅
1:17
adithya_s_k's profile picture

Introducing Omnidocs - Unified Visual Document processing toolkit ✅ 6+ document processing tasks ✅ 15+ Model supported ✅ Runs on consumer GPU/ Mac ✅ Supports multiple backend - Pytorch, vllm, mlx, open responses API Get started → pip install omnidocs✅

Adithya S K

22,243 views • 8 months ago

Ever felt lost in your own codebase ? 👉🏻 Meet Gitvizz - instantly turn your codebase into interactive graphs so you can see how everything’s wired Check it out at gitvizz[.]com
0:49
adithya_s_k's profile picture

Ever felt lost in your own codebase ? 👉🏻 Meet Gitvizz - instantly turn your codebase into interactive graphs so you can see how everything’s wired Check it out at gitvizz[.]com

Adithya S K

24,206 views • 1 year ago

Here is quick tutorial on the quickest way to repoduce/run NanoChat with minimal GPU setup time using Modal
2:11
adithya_s_k's profile picture

Here is quick tutorial on the quickest way to repoduce/run NanoChat with minimal GPU setup time using Modal

Adithya S K

17,996 views • 1 year ago

No more content to load