Ripu

Exploring the frontiers of AI, Generative AI, Robotics, and Deep Tech, driving the transformations that are reshaping who we are and what we will become.

[Ripu @Ripuhiring]

@Kapil_Jyani_ Dear Kamlesh,
You need to know AI a little bit more, it's now all about bots anymore bro

[Ripu @Ripuhiring]

One of the simplest explanations of AI Agents I've heard comes from Jensen Huang.

Think of an AI agent as a worker in a workshop.

The model is the brain that thinks and reasons.

The harness gives structure and orchestrates the work.

Tools and skills allow the agent to interact with the world.

The runtime is the workshop where the work actually gets done.

This analogy cuts through much of the hype around Agentic AI.

[Ripu @Ripuhiring]

What an innovative way to catch the brands core messaging through the journey ...connecting to nature and self

[Ripu @Ripuhiring]

@AMISHDEVGAN @RahulGandhi सब ठीक है, यह संतुलन बनाने की कोशिश क्यों? 🤣🤣🤣

[Ripu @Ripuhiring]

The problem is that technology is not governed by protectionism; it is governed by excellence. In tech, you either compete with the best or get left behind. The only real winner is "the best"—everyone else eventually fades away.

India still spends too much time protecting average tech companies and startups instead of going all-in on innovation. We are surrounded by businesses that simply imitate original products and survive behind policy shields.

Let's be honest: how many of these companies have seriously invested in R&D or funded meaningful research at our premier engineering institutes?

Instead of asking for more protection, face the competition, invest in innovation, and earn your place on merit.

[Ripu @Ripuhiring]

Ship traffic in the Strait of Hormuz now

[Ripu @Ripuhiring]

@join2manish AI isn't replacing intelligence—it's exposing it

[Ripu @Ripuhiring]

AI isn't replacing intelligence—it's exposing it. Repetitive skills were often mistaken for intelligence. The real advantage now is human judgment, creativity, and original thinking.

[Ripu @Ripuhiring]

True, Vijay. Now the competition will be about genuine human intelligence, not repetitive-task skills that were often mistaken for intelligence. We are moving to the next level, one where real creativity, judgment, reasoning, and originality can be recognized and valued without flukes or shortcuts.

[Ripu @Ripuhiring]

So, did you raise the same concern with @telegram or @durov and ask them for clarification?

(I am tagging them to give their counter on this)

Do you really think someone who has never built even a simple world-class app is in a position to confidently criticize an app that serves more than 150 million users in India alone?

I'm not saying your concerns are wrong. Criticism is important. But a one-sided demo without context often reflects insecurity more than brilliance. Meaningful analysis requires looking at both strengths and weaknesses, not just highlighting what supports a predetermined narrative.

[Ripu @Ripuhiring]

@ANI But why

[Ripu @Ripuhiring]

@SudhanshuTrived @JagranNews @narendramodi @BJP4India @BJPCentralMedia @BJP4UP @PMOIndia @NitinNabin @JPNadda @AmitShah @myogiadityanath @mppchaudhary अपुन लिख के देता है, नोट कर लो।

[Ripu @Ripuhiring]

Yup. If you want high talent density for a high-level product, you need to continuously build a culture where exceptionally talented people don't lose sight of the mission, their motivation, or their zeal to work with you. Talent density isn't just about hiring great people—it's about creating an environment that keeps them engaged, challenged, and aligned with a larger purpose.

[Ripu @Ripuhiring]

@gharkekalesh ये सब तो सामान्य बातें हैं, आप लोग ज़्यादा उत्साहित मत होइए। कालीन और कबूतर पर ही ध्यान केंद्रित रखिए

[Ripu @Ripuhiring]

लेकिन टोल वाले कहाँ हैं, जिनका पूरा काम ही यह सुनिश्चित करना है कि एक्सप्रेसवे सुचारु रूप से चले और उसका रखरखाव ठीक से हो? यह मत कहिए कि उनके पास ऐसी परिस्थितियों में एक्सप्रेसवे साफ़ करने की मशीनें तक नहीं हैं। अगर ऐसा है, तो फिर वे सिर्फ़ टोल वसूलने और पैसा लेने के लिए ही हैं, काम करने के लिए नहीं?

[Ripu @Ripuhiring]

@govindprataps12 @grok दादा, ओई तो मिथुन होबे।

[Ripu @Ripuhiring]

The honeymoon phase of "unlimited AI for everyone" is officially over. 📉

According to a recent Business Insider report, corporate tech is experiencing massive sticker shock as employees lean heavily into high-context models and complex agentic workflows. Enterprises are rapidly shifting from permissive AI adoption to strict token limits and internal spending controls.

The numbers are eye-opening:

Uber reportedly exhausted its entire planned 2026 AI coding budget in just the first four months of the year.

Meta recently issued an internal memo to thousands of employees tightening oversight on token usage, with internal AI costs on track to hit billions.

Pylon (an enterprise software startup) had to set token ceilings for non-technical staff after realizing their scaled Anthropic plan was pacing toward a staggering $1.4 million bill.

The Pivot from "Tokenmaxxing" to "Tokenminimizing"For the last year, companies built internal leaderboards and incentivized employees to maximize AI use. But treating token consumption as the primary metric created a classic Goodhart’s Law trap: users inflated scores by running low-value tasks through background agents, skyrocketing compute costs with no clear link to shipped value.
Now, the pendulum is swinging hard toward governance.

What this means for software engineers and leaders:

The End of Unlimited Context: Running massive repositories or long documents through prompts without optimization is becoming a financial liability.

From Seat Licenses to Metered Budgets: Finance teams are treating AI less like SaaS and more like the electric utility company. Expect to see per-user token caps and approval workflows soon.

The Rise of Cost-Aware Design: Engineers who know how to optimize prompts, use smaller fine-tuned models, or implement context caching will be highly valued.

The solution isn't to clamp down on AI completely—it’s moving away from a simple "on/off switch" and moving toward robust governance layers that track output value, not just token input.

How is your organization handling AI budget caps? Are you starting to feel the "token squeeze"? 👇

https://t.co/OJxwD3c7ej

#ArtificialIntelligence #SoftwareEngineering #TechBudgets #GenerativeAI #FinOps #TechLeadership

[Ripu @Ripuhiring]

https://t.co/dVjlRzujA9

[Ripu @Ripuhiring]

https://t.co/RF5NGi0D82

[Anjney Midha @AnjneyMidha]

last evening, @DavidSHolz reminded a lot of us in technology why we joined the industry