Murali Manohar

AI something @AlephAlpha. Post training models.

Built agentic chat & evals

[Murali Manohar @gitlostmurali]

@tugot17 @UniofOxford Congrats! 👏

[Murali Manohar @gitlostmurali]

Ending this year with a blog on RL environments:

https://t.co/wfGtEormJy

Talks about reward hacking, sandboxing, curriculum learning, tool calling - all the stuff that can break when you actually try to train agents

[Piotr Mazurek @tugot17]

RL is cool, but what do you actually need to know about hardware and infra to predict its future? Check out our new piece on tensoreconomics:

[Murali Manohar @gitlostmurali]

Here's my blog on Understanding GRPO:

https://t.co/r8KTFLYz2L

[Murali Manohar @gitlostmurali]

@tanay_mehta Nice nice

[Murali Manohar @gitlostmurali]

@rishdotuk Six of crows if you haven't started it

[Murali Manohar @gitlostmurali]

Thanks!

https://t.co/cVZAPeYJDQ
It's a countdown task - "Use every given number exactly once to build an equation that hits the target."

As a next step, I wanted the small model to decide if it has to forward the equation or solve on its own. I introduced latency as a negative reward but it didn't work quite well. And then I went to other projects.

[Murali Manohar @gitlostmurali]

An interesting thing happened when the remote model is swapped with a bad quality 7B LLM. After failing to get the remote model to provide right equations, the local model started providing hints in its prompt for the remote model. In a few instances, the local model started solving the problem and just passed the equation in remote model's prompt for the remote model to just **forward** the equation.

[Murali Manohar @gitlostmurali]

@brendanh0gan Sounds similar to the experiment I tried a few months ago. I tried it with qwen-3b and o3-mini

Spoiler: It works!
https://t.co/C99Esn1D9n

[Murali Manohar @gitlostmurali]

@tugot17 Amazing! 🚀

[Murali Manohar @gitlostmurali]

@rishdotuk Ouch! I like the personality.

Imagine having this as the PR reviewer before one submits it to the team

[Murali Manohar @gitlostmurali]

@goyal__pramod Always. Visualizations are rewarded since humans are visual creatures.

To speed this process, you can try these:

1. Claude renders diagrams with Artifacts feature.
2. Try asking AI for mermaid versions and later manually convert to excalidraw or ilk

[Murali Manohar @gitlostmurali]

@sh_reya @HamelHusain Thank you! I'll follow up over email to discuss the budget and see what can work. Appreciate the flexibility 🙏
Thanks again!

[Murali Manohar @gitlostmurali]

Given @HamelHusain and @sh_reya 's credibility, I would love to take the evals course. But 2000$ seems like a lot. Even to convince the corporate. What are some best alternatives here?

[Hamel Husain @HamelHusain]

Early bird discount ends next Friday: https://t.co/cg6BtHZBF5

[Murali Manohar @gitlostmurali]

@tanay_mehta @doSwayamExist Ouch. How pathetic!! This incident and intern-shaming Apple's latest paper is deeply concerning. This is classism in a different form.

This can only change if indian teachers/ professors are "okay" with removing Sir/Madam. And how cool is it to say, Prof. Sanyal/Prof. Chawla?!

[Murali Manohar @gitlostmurali]

Love how Sarvam placed their announcements!!!
Like here's a model we trained to here's a product we built

[Sarvam @SarvamAI]

Introducing Sarvam Samvaad 🚀

Sarvam Samvaad, our Conversational AI Platform, is designed to help enterprises build, test, and launch AI agents fluent in 11 Indian languages.

▪️Power interactions across telephone, WhatsApp, web, and apps
▪️Handle complex phrases, alphanumerics, and proper nouns with precision
▪️Listen in on every conversation and discover deep insight
▪️Scale your support with pricing built for India

Trusted by leading brands across the country, the platform delivers production-ready agents so enterprises can move from pilot to full-scale deployment in just days.

It’s time to level up customer experience with Sarvam Samvaad.

Watch it in action ↓

[Murali Manohar @gitlostmurali]

@sama @tanay_mehta plus users too

[Murali Manohar @gitlostmurali]

The linter is still complaining. Let me try a different approach by adding type ignores

Claude models (3.7 & 4) are ruthless

[Murali Manohar @gitlostmurali]

@t_blom I came across folks who said:

"If you find the clients and take care of the tech stuff, you'll be the co-founder". How dependable can a person be!!?

[Murali Manohar @gitlostmurali]

Cursor Code is exactly what I want my cursor to be. Precise edits and incremental.