Prateek Pandey

Distributed Systems & Opensource Enthusaist | Gopher | Containers | @kubernetesio contributor | Maintainer @kyverno @OpenEBS | Organiser @DevOpsDaysIN

[Anthropic @AnthropicAI]

Claude Code can now connect to remote MCP servers.

Pull context from your tools directly into Claude Code with no local setup required.

[Theodora Chu @chu_onthis]

new mcp spec just dropped:
1. auth is fixed! at last!
2. elicitation now makes it possible for a server to ask an end-user for more info, enabling more agentic behaviors
3. structured tool outputs makes it easier to reason about tool responses
4. more security documentation

[tobi lutke @tobi]

I really like the term “context engineering” over prompt engineering.

It describes the core skill better: the art of providing all the context for the task to be plausibly solvable by the LLM.

[Andrej Karpathy @karpathy]

Part 2 of this mystery. Spotted on reddit.
In my test not 100% reproducible but still quite reproducible.
🤔

[Balaji @balajis]

AI PROMPTING → AI VERIFYING

AI prompting scales, because prompting is just typing.

But AI verifying doesn’t scale, because verifying AI output involves much more than just typing.

Sometimes you can verify by eye, which is why AI is great for frontend, images, and video. But for anything subtle, you need to read the code or text deeply — and that means knowing the topic well enough to correct the AI.

Researchers are well aware of this, which is why there’s so much work on evals and hallucination.

However, the concept of verification as the bottleneck for AI users is under-discussed. Yes, you can try formal verification, or critic models where one AI checks another, or other techniques. But to even be aware of the issue as a first class problem is half the battle.

For users: AI verifying is as important as AI prompting.

[Raj Kunkolienkar @kunksed]

We all want to root for Sarvam. But wanting isn't enough when you're selling duty and technical jargon instead of desire. It’s a narrative problem.

Right now, my perception of Sarvam's unsaid positioning feels like "We’ve done some cool technical stuff. Support us because we're Indian AI."

A semi technical idiot such as I can’t make head or tail of what they’re claiming to have achieved. The latter bit is also the exact narrative trap that didn’t end up helping Koo, Hike, and every other "patriotic alternative."

In 2025, when every other day brings a new AI breakthrough, you can't ship a model and expect the world to care just because it handles Indic languages. Even if you build something revolutionary tomorrow, adoption needs more than good intentions and GitHub stars.

The "we're going B2B" defense doesn't hold either. Today's enterprise buyers discover through consumer buzz. The lines have blurred beyond recognition.

Here's the playbook that I, a technical dumbo thinks might actually work?

1. Make magic, not models

Nobody (aka semi technical idiots like me) cares about your 24B parameters. They probably care about their grandmother finally being able to use technology the way they do. Show that. Ship a demo that makes someone cry because their regional poetry finally sounds beautiful in AI. Or whatever it is that you can pull off. It’s an emotional game. Don’t wait on me to figure out what you’re capable of.

2. Own the conversation

Where are your founders showing up? Your engineers? In 2025, faceless companies die. Perplexity has Aravind. OpenAI has Sam. Sarvam has... a Twitter account and a blog? Get your smartest people on camera daily. Make them influencers. Yes, it's ‘cringe’. Do it anyway.

3. Buy the hype you deserve

Those "organic" Silicon Valley hype cycles? You already know they cost millions. Stop being precious. Deploy ₹1Cr on strategic amplification. Seed Twitter. Pay influencers. Create FOMO. The best tech with the worst distribution equals failure.

4. Pick a fight

"Big AI doesn't understand when your mom mixes three languages in one sentence. We do." That's a story. "Support Indian AI" isn't. Make Google the villain. Make context-blindness the enemy. Give people something to rally against. Some bait. Please.

5. Ship experiences, not infrastructure

One viral consumer moment drives more enterprise leads than hundred cold emails. Ask Jasper. Ask Notion. Ask any AI company that matters. Or ask Ghibli.

Stop asking for support. Start demanding attention. Show us one thing — just ONE thing — that makes us go "Holy shit, only sarvam can do this."

Do that, and we won't just rally behind you. We'll evangelize for you.

Help us visualize why this matters to our lives, not our national pride or something. The market of dumbos like me is waiting.

[Prateek Pandey @imPrateek14]

Finally no pin 📌 to create local copies of variables in for loops and enhanced routing pattern

[Go @golang]

🥳 Go 1.22.0 is released!

🗒️ Release notes: https://t.co/WGgf1Nm0Mu

📦 Download: https://t.co/pRCJTZElxC

#golang

[CNCF @CloudNativeFdn]

After nearly 10 years of #Kubernetes, @thockin looks at the challenges and opportunities for the next decade!

Where will AI/ML lead us? How do we make running multi-cluster easier? How do we fix complexity issues?

Check in at #KubeCon 2033?

[Prateek Pandey @imPrateek14]

Anmol Virmani and Palash Goel from @zomato presenting Zomato's migration to @ClickHouseDB have truly redefined their logging capabilities, all while unlocking substantial cost savings.

[Prateek Pandey @imPrateek14]

@devopsdays_blr day 2 @neependra and Abdul Karim demonstrating how we can use the eBPF tools to visualise different security aspects for a Kubernetes Cluster.
#k8s #security with #eBPF
#devopsdaysblr

[Floor - oh I'm gone @DevOpsBarbie]

@devopsdays_blr Anchan Jain & Rajat Porwal at @devopsdays_blr with a walk-through of cloud scaling and service mesh intricacies, offering us (or: DevOps professionals) a guide for facing unexpected challenges #devopsdays

[Prateek Pandey @imPrateek14]

@_ashish_tiwari from @elastic talking about #Elastic + #GenAI @devopsdays_blr #devopsdays

[DevOpsDays Bengaluru @devopsdays_blr]

We have @_ashish_tiwari from @elastic telling us more about Elastic, applications, use cases

Thank you @elastic for being our sponsor. #DevopsDaysBLR is possible because of you 💖

[DevOpsDays Bengaluru @devopsdays_blr]

We have Ranbir & Saad talking about
Empowering Developers, Uniting DevOps: Backstage - Your Gateway to Seamless Collaboration 🎉 #Backstage #DevOps #DevOpsDaysBLR

[DevOpsDays Bengaluru @devopsdays_blr]

📢 Important Update for #DevOpsDaysBLR!

Due to a few logistical challenges at our current venue we have had to make a difficult decision to change the venue location within Bengaluru.

We remain committed to delivering an exceptional event 💖

#DevOpsDays #DevOpsDaysBLR

[DevOpsDays Bengaluru @devopsdays_blr]

We're still looking for sponsors to make #DevOpsDaysBLR 2023 happen this year.

Let's make this event a huge success together💖

Check out the sponsorship details here 👇🏼👇🏼
https://t.co/k973gUZrTo

#DevOpsDays #DevOpsDaysBLR #DevOps #Reliability #DevOpsCulture

[Harsha Bhogle @bhogleharsha]

Do try and listen to the early episodes of #MissionISRO to get an idea of the audacity, the brilliance and the intense patriotism of our scientists in the 50s and 60s when we were a nation forced into poverty by colonial plunder. The philosophy of doing a lot with few resources and great ingenuity was established then. And a series of great scientists built on that philosophy to create history on this budget. @isro.

[ISRO @isro]

Chandrayaan-3 Mission:
'India🇮🇳,
I reached my destination
and you too!'
: Chandrayaan-3

Chandrayaan-3 has successfully
soft-landed on the moon 🌖!.

Congratulations, India🇮🇳!

#Chandrayaan_3
#Ch3

[Prateek Pandey @imPrateek14]

Touching the moon, one step at a time 🌝 India's Chandrayaan 3 successfully lands on the lunar surface, making another giant leap for space exploration! 🇮🇳🚀 #Chandrayaan3Success #IndiaOnTheMoon

[DevOpsDays Bengaluru @devopsdays_blr]

🚨 4 Days to go to submit your proposals at #DevOpsDaysBLR 2023 !!

Hurry, the clock is ticking!