Raghava Reddy

Die with Memories not with Dreams✌️.
Entrepreneur, Software Developer and Trader📈☂️

[Raghava Reddy @raghava_rdy]

what the hell with the @IndiGo6E Both business and support are worse than expected. A delayed flight is understandable, but the lack of smart support is frustrating.

@IndiGo6E should be held accountable and accommodate those who suffering a 12-hour delay.

[Raghava Reddy @raghava_rdy]

AI agents, real terminal UI, team orchestration, nested sessions, and Claude Code—all from your browser.

Run Agent Maestro locally, manage multiple AI teammates, and build without workflow chaos.

GitHub: https://t.co/q7V6jnJVag

bash scripts/start-web.sh

#AI #OpenSource #Agent

[Vivo @vivoplt]

15 AI related accounts you should follow on Twitter:

1. @karpathy
2. @fchollet
3. @ylecun
4. @AndrewYNg
5 @rasbt
6. @dair_ai
7. @lilianweng
8. @jeremyphoward
9. @simonw
10. @_akhaliq
11. @ID_AA_Carmack
12. @gwern
13. @goodside
14 @drfeifei
15 @demishassabis

Let me know who I missed guys

[Matt Pocock @mattpocockuk]

What is /grill-me?

Here:

https://t.co/CFbsSren8D

[Raghava Reddy @raghava_rdy]

@airtelindia is looting

Not only airtel, Jio also same my mother won't use much but still guys are going to extreme extent to loot public paise

[Chris Tate @ctatedev]

Portless killed :3000

Dev servers got stable names like myapp.localhost

Agents could use worktrees in parallel without stepping on each other

Now it's easier than ever in v0.11

Just run: portless

Zero config. Zero args. Zero code changes.

[Raghava Reddy @raghava_rdy]

I left my laptop in fridge for 4 hours and enabled remote control through @claudeai it's just cooked for me what ever I requested.

[Akshay 🚀 @akshay_pachaar]

https://t.co/LLvSoQyQaP

[Aksel @akseljoonas]

Introducing ml-intern, the agent that just automated the post-training team @huggingface

It's an open-source implementation of the real research loop that our ML researchers do every day. You give it a prompt, it researches papers, goes through citations, implements ideas in GPU sandboxes, iterates and builds deeply research-backed models for any use case. All built on the Hugging Face ecosystem.

It can pull off crazy things:

We made it train the best model for scientific reasoning. It went through citations from the official benchmark paper. Found OpenScience and NemoTron-CrossThink, added 7 difficulty-filtered dataset variants from ARC/SciQ/MMLU, and ran 12 SFT runs on Qwen3-1.7B. This pushed the score 10% → 32% on GPQA in under 10h. Claude Code's best: 22.99%.

In healthcare settings it inspected available datasets, concluded they were too low quality, and wrote a script to generate 1100 synthetic data points from scratch for emergencies, hedging, multilingual etc. Then upsampled 50x for training. Beat Codex on HealthBench by 60%.

For competitive mathematics, it wrote a full GRPO script, launched training with A100 GPUs on https://t.co/udm7xGpNzR, watched rewards claim and then collapse, and ran ablations until it succeeded. All fully backed by papers, autonomously.

How it works?

ml-intern makes full use of the HF ecosystem:
- finds papers on arxiv and https://t.co/brvCC7fLPa, reads them fully, walks citation graphs, pulls datasets referenced in methodology sections and on https://t.co/hrJuRkRyzi
- browses the Hub, reads recent docs, inspects datasets and reformats them before training so it doesn't waste GPU hours on bad data
- launches training jobs on HF Jobs if no local GPUs are available, monitors runs, reads its own eval outputs, diagnoses failures, retrains

ml-intern deeply embodies how researchers work and think. It knows how data should look like and what good models feel like.

Releasing it today as a CLI and a web app you can use from your phone/desktop.
CLI: https://t.co/l3K1PslZ1n
Web + mobile: https://t.co/orko5srL4H

And the best part? We also provisioned 1k$ GPU resources and Anthropic credits for the quickest among you to use.

[Turing Post @TheTuringPost]

Must-read research of the week

▪️ Dive into Claude Code: The design space of today’s and future AI agent systems
▪️ Lightning OPD: Efficient Post-Training for LRMs with Offline On-Policy Distillation
▪️ Self-Distillation Zero: Self-Revision Turns Binary Rewards into Dense Supervision
▪️ TIP: Token Importance in On-Policy Distillation
▪️ Toward Autonomous Long-Horizon Engineering for ML Research
▪️ Memory Transfer Learning: How Memories are Transferred Across Domains in Coding Agents
▪️ Agentic Aggregation for Parallel Scaling of Long-Horizon Agentic Tasks
▪️ Introspective Diffusion Language Models
▪️ Maximal Brain Damage Without Data Or Optimization: Disrupting Neural Networks Via Sign-Bit Flips
▪️ The Past Is Not Past: Memory-Enhanced Dynamic Reward Shaping
▪️ KnowRL
▪️ Rethinking On-Policy Distillation of LLMs
▪️ Continuous Adversarial Flow Models

Find all the links and the most important AI news of the week here: https://t.co/sIqDJ3Dp4p

[Ajit kumar @ajitcodes]

Learn AI for free directly from top companies.

1 - Anthropic:
https://t.co/1M15GC7cOY

2 - Google:
https://t.co/AIkLM6XjVL

3 - Meta:
https://t.co/UWdSCK5daz

4 - NVIDIA:
https://t.co/B458g8uF1u

5 - Microsoft:
https://t.co/UChXyK2ZQh

6 - OpenAI:
https://t.co/Do088drBz6

7 - IBM:
https://t.co/4XKRjVNMKA

8 - AWS:
https://t.co/36WQ7H0vfT

9 - https://t.co/saSYDpvXFf:
https://t.co/mtt2yYxm6a

10 - Hugging Face:
https://t.co/YYHwWDpwPF

👇Comment "Learning" if you find this helpful.

Repost so others can take help.

Must bookmark for future reference.

[Eric ⚡️ Building... @outsource_]

🚀 NEW GEMMA 4 31B TURBO DROPPED

Runs on a SINGLE RTX 5090:

⚡️18.5 GB VRAM only (68% smaller)
🧠51 tok/s single decode
💻1,244 tok/s batched
🤖15,359 tok/s prefill ← yes, fifteen thousand

🚨2.5× faster than base model with basically zero quality loss.

It hits Sonnet-4.5 level on hard classification tasks…
at 1/600th the cost.

Local models are shipping faster than we can test 👇🏻

🔥 HF: https://t.co/XUvVZBj9AX

[mitsuri @0xmitsurii]

How was the show Silicon Valley so ahead of its time?

[Raghava Reddy @raghava_rdy]

What the fuck whatsapp started with ads.

I guess it's high time to build something like WhatsApp

[Raghava Reddy @raghava_rdy]

@mert Where money exists, encryption is required.
Where money doesn’t, freedom remains.

[Raghava Reddy @raghava_rdy]

Fall in love with the process rather than the result. You will be more satisfied.

[sysls @systematicls]

https://t.co/FQe5bCBqW1

[Raghava Reddy @raghava_rdy]

@codewithDiyaa It works initially, but then fails to perform as expected.

[Raghava Reddy @raghava_rdy]

@ick_real life

[Raghava Reddy @raghava_rdy]

Brutally honest truth:

Ideas are cheap. Numbers & Execution decide survival.