Indrajit Bhosale | इंद्रजित भोसले

Computer Systems | History | GGMU | Running (In this particular order)

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

My time at @CarnegieMellon @SCSatCMU and @inicmu ends :)
Special thanks to @_ritvika_ @BhosalePratim and @ShivaniRB16 for their support throughout one of the most challenging phases on my life.
I already miss the grind :)

[Han Xiao @hxiao]

SGLang and vLLM are racing 😭. There’s no weekend for inference guys.

[vLLM @vllm_project]

vLLM matches DeepInfer, DeepSeek's own engine for DeepSeek-V4.1-Flash, on DeepSeek's decode configuration across batch sizes 8 to 256, and does it at DEP32 against DeepInfer’s EP64DP64.

DeepInfer was originally built on vLLM. By integrating DeepSeek's open-source kernels and extensively optimizing prefill and decode, vLLM now matches DeepInfer's performance curves.

[𝘄𝗮𝗻𝗱𝗲𝗿𝘁𝗿𝗮𝘃𝗲𝗹𝘀 @visualsofearth1]

Pacific Northwest

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

@aakashb_95 😞

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

@KranenKyle @mahe

[Kyle Kranen @KranenKyle]

Super cool to see the community interest in DeepInfer support for Dynamo!

https://t.co/gROlcyK85F
https://t.co/SOd0bM7AIQ

We have a PR with disagg support we hope to drop tomorrow, so if you'd like to collaborate, join us on slack!
https://t.co/yB81j4ViMW

[sq @stonksqween]

How does one even think of this

[Liam Canning @LiamPaulCanning]

Only reason I want Conte is for him to kill the players on the training field. They deserve it

[Pratim Bhosale @BhosalePratim]

I built an iOS Gym app that counts my reps and corrects my stance.

It runs offline and generates audio on the go on my phone. No API costs.

This way, I can focus purely on my breathwork and not worry about counting reps or technique.

Built this with Phonon.
Open-source on-device TTS model from @GradiumAI

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

@KranenKyle Thanks Kyle! :)

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

Day0 support with Nvidia Dynamo.
This was a fun collaboration 🤝

[NVIDIA AI @NVIDIAAI]

And @AIatMeta went live with Muse Spark Open!

The 816B model was trained on NVIDIA GB300 NVL72. Its 1M-token context window gives coding agents room to work across large codebases.

Meta also produced NVFP4 checkpoints optimized for Blackwell Ultra, delivering over 50K tokens/sec/GPU to improve throughput and lower cost per token.

Congrats to the @Muse team.

More in our blog: https://t.co/WcSFpdHowL
NVFP4: https://t.co/gBvebf8XxE

[M I K E ©️ @mike_muhima]

Funny how he purchased a defender for himself but not for the team.

[NVIDIA AI @NVIDIAAI]

Meet NVIDIA Nemotron 3.5 Super VL

Our new open model combines reasoning and tool use with text, image, document, and video understanding for long-running AI agents.

120B total parameters with 12B active 🧵

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

Cisco has this since idk 2 decades or more.
Not sure why are people soo surprised.
A lot of companies have DevX orgs that routinely does this.
Cisco even had/has their own Jira, ServiceNow etc.

[Anthropic @AnthropicAI]

New on our Engineering blog: Creating in-house source control.

In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale.

Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: https://t.co/wzNkwHs7IN

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

Part of a few!
Stay tuned!

[NVIDIA AI @NVIDIAAI]

So many models launching this week!
And we have one of our own coming...⌛️

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

@s6juncheng @sachaarbonel

[Indrajit Bhosale | इंद्रजित भोसले @devnull_0]

@svembu Sir this is misleading, the outlook for software is NOT uncertain for your competitors.
MSFT, Google, Salesforce, Adobe are all posting record revenue and profits. Maybe it's uncertain for Zoho right now, which is understandable.
Every company goes through ups and downs.

[Hina Khan @cesthinakhan]

Two grifters exchanging notes on monetizing performative activism

[S|(ene 🍭🇮🇳 @mysteriou_s1]

MFs sabotaged BSNL to favour Ambani, and now they’re talking about nationalism. 🤣