Chetan Yewale

Cloud consultant at AWS. Other interests - history, science, science fiction and fintech.
Views are my own

[Chetan Yewale @3B0dyProblem]

@sivalabs Create a new spec which references previous spec - that creates a trail/link of specs when you want to change a feature or enhance it

[Chetan Yewale @3B0dyProblem]

Would suggest you recommend this blog article from Marc Brooker (distinguished engineer at Amazon ) : https://t.co/2A2Vk5n8av

Most people misunderstand any new process (especially with a new paradigm ).

SDD is the best way for AI assisted development (note that i dont say its perfect, human judgement and review plays an important role ). But its definitely better than vibe coding , works great for large projects. We are using this technique for many projects and its working fine for enterprise projects in my experience.

[Chetan Yewale @3B0dyProblem]

Had some thoughts about the relation between Lord Rahu and AI, combined those into a small article

AI isn’t just a tech revolution; it is the modern manifestation of **Rahu**. 🪐🤖
In Vedic mythology, Rahu is the smart Asura—the master of material gain, unconventional technology, and total disruption. When you get addicted to "vibe coding," you see Rahu’s influence: the feeling of infinite possibility and the urge to never stop building.
But this "Amrut" comes with a price:
* **Speed over Depth:** Just like Rahu, AI prioritizes rapid output over deep expertise. It creates the illusion that we can build complex systems without the years of mastery traditionally required.
* **The Great Divider:** It’s a tool that makes the rich richer. While it offers "material" success, the high cost of subscriptions and compute power creates a new divide for those who can’t afford the entry fee.
* **The Shadow Side:** Rahu rules over illusions and fakery. We see this in the dark side of AI—deepfakes, digital deception, and the blurring of reality.
* **Challenging Wisdom:** From medicine to law, AI is causing us to question traditional authority. While it democratizes simple tasks (like basic medical advice), it risks eroding our respect for human experience and the nuance of experts.
* **The Environmental Cost:** Rahu's hunger is insatiable. The massive energy costs of AI are an indirect harm we are only beginning to calculate.
Ultimately, Rahu pretended to be a god to taste the nectar of immortality. Modern AI does the same—it doesn't "think," but it mimics the **illusion of intelligence** so perfectly that it has successfully disrupted the very fabric of our traditional world.
The disruption is here. The question is, can we handle the Asura we’ve invited into the room? 🎭⚡️

[Chetan Yewale @3B0dyProblem]

AI isn’t just a tech revolution; it is the modern manifestation of **Rahu**. 🪐🤖
In Vedic mythology, Rahu is the smart Asura—the master of material gain, unconventional technology, and total disruption. When you get addicted to "vibe coding," you see Rahu’s influence: the feeling of infinite possibility and the urge to never stop building.
But this "Amrut" comes with a price:
* **Speed over Depth:** Just like Rahu, AI prioritizes rapid output over deep expertise. It creates the illusion that we can build complex systems without the years of mastery traditionally required.
* **The Great Divider:** It’s a tool that makes the rich richer. While it offers "material" success, the high cost of subscriptions and compute power creates a new divide for those who can’t afford the entry fee.
* **The Shadow Side:** Rahu rules over illusions and fakery. We see this in the dark side of AI—deepfakes, digital deception, and the blurring of reality.
* **Challenging Wisdom:** From medicine to law, AI is causing us to question traditional authority. While it democratizes simple tasks (like basic medical advice), it risks eroding our respect for human experience and the nuance of experts.
* **The Environmental Cost:** Rahu's hunger is insatiable. The massive energy costs of AI are an indirect harm we are only beginning to calculate.
Ultimately, Rahu pretended to be a god to taste the nectar of immortality. Modern AI does the same—it doesn't "think," but it mimics the **illusion of intelligence** so perfectly that it has successfully disrupted the very fabric of our traditional world.
The disruption is here. The question is, can we handle the Asura we’ve invited into the room? 🎭⚡️

[Chetan Yewale @3B0dyProblem]

That has not been my experience.not sure about other teams .

In our team - yes we are expected to mention how much productivity gains happened per project or our daily work , and official tracking by Kiro badges has started recently in past 2-3 months . But officially I don’t know of anyone in our team who was penalized for not using AI. It’s encouraged for sure and we are supposed to provide metrics (honest metrics) on how much AI we used and how productivity gains we think happened . But penalizing I don’t know of any case

[Chetan Yewale @3B0dyProblem]

@whynesspower How can that be true ? Kiro rollout to all employees happened these last 2-3 months ago only . Infact Kiro was released for internal testing only in July 2025 . Before there was Amazon Q CLI and plugin . So this will be the first time that AI usage will get considered .

[Chetan Yewale @3B0dyProblem]

Java interfaces and default methods in interfaces exist for this reason . Rigid inheritance hierarchies are well known antipattern since a decade or so in Java and .NET world.
While approach of Go maybe flexible , it’s better readability if I can clearly see which interfaces a class is implementing.

Good programmers with an eye for quality code don’t create these rigid hierarchies and neither the interfaces upfront. You keep working , and if you see that there are different implementations but with similar patterns and you see this pattern atleast 3 times , that’s when you think of creating an interface or applying a design pattern if it makes sense.

[Chetan Yewale @3B0dyProblem]

@svpino I created a Claude skill for reading and writing PDF files.

[ThePrimeagen @ThePrimeagen]

I have been thinking about this a lot.

I think for a great many of engineers, the ones who did it because they loved it only to discover that money was in fact at the end of the rainbow found both the journey and the destination satisfying. In fact, I think I can argue with authority that the destination was only satisfying as the journey was difficult. The hard-fought evenings spent toiling away on an idea and codebase that slowly gives way to your vision was an incredible experience.

The group of people that fell into this category of hard-fought journey and destination we will call them tinkerers. One thing tinkerers have always hated is the already known problems. The journey is clear as day. The obstacles minor inconveniences. Its purely a matter of typing the solution into the terminal. This is also why I think so many of this group goes out and does open source, or starts companies. Work largely falls into this category with few exceptions.

From this reason is why I largely find UI work soul sucking. I know the solution, its a matter of just looking up the details and putting it into my editor. yawn. CSS, flex box this, grid that, put the tailwind classes in the bag.

To me, the LLM software world is with little to no journey and discovery. Its more of simply taking my high level idea and just formulating it into testable, atomic chunks that can be verified. I have traded my favorite part, discovery and raw creation, with itemized list of TODOs and patience and "No Mistakes."

To this, every morning from 6 to 9 I simply just hand code every thin. even UI things. It is because I want journey and discovery and raw creation. Maybe one day comes and its just so futile that I stop this. But for now, I still see such great value in this. I see such better thought through products. Because slowing down and truly thinking through everything. The architecture, the design, everything is an expression of discovery and creation. And I love it.

I am sure there will come a day, maybe even in the next 6 months where I change my mind. For now, I pursue the love of the game intentionally.

I do also believe that there exists people who get the same joy I got from building with tears and sweat by prompting LLMs. I am positive of it. I just don't understand how. But people love UI work. I also don't understand that.

[Chetan Yewale @3B0dyProblem]

Today saw a doctor using Ubuntu OS . Glad to see adoption of Ubuntu , a free Linux OS . Ubuntu has long been user friendly Linux OS and most common software is already there . And people who just use the browser for most of their work do not really need windows specific software .

[Chetan Yewale @3B0dyProblem]

Even TCS employees are using Copilot with Enterprise version . There are few projects where the SoW(statement of work) does not allow usage of AI coding assistants. But most projects are allowing specific AI tools . Now companies are ensuring that in SoW the usage of AI is allowed for development

[Chetan Yewale @3B0dyProblem]

Even TCS employees are using Copilot with Enterprise version . There are few projects where the SoW(statement of work) does not allow usage of AI coding assistants. But most projects are allowing specific AI tools . Now companies are ensuring that in SoW the usage of AI is allowed for development

[Chetan Yewale @3B0dyProblem]

While Sarvam has done a great job, I want to point a few things:
1. Gemini Live 2.5 flash is available in Mumbai region now (since December 2025). So no your data does not go to other regions if you are using appropriate region
2. Gemini Live is a speech to speech model. Per my analysis Sarvam AI is not speech to speech .
What’s the difference ? In case of Sarvam , the speech is converted to text first , AI processes the text and returns response back as speech.

That’s not what happens for Gemini live. It directly understands the speech and responds in speech. The speech to speech models are very costly to build and they take this direct approach to avoid extra step in this pipeline

[Chetan Yewale @3B0dyProblem]

Built a useful vibe coded tool for myself in 2 hours(using Claude code) .. I can use Claude code from my WhatsApp by messaging myself .. Claude code sits on a VPS (can be your raspberry pi too at home , just needs to be reachable from your phone )

It works. Able to browse files, chat with Claude code , take actions (CC when asks confirmation to edit file )

Created a git repo
https://t.co/mXqJTdkcUh

I was seeing what openclaw and tinyclaw are doing .. thought of using tinyclaw but realized my requirements is simple - to do vibe coding from my phone and avoid always opening the laptop for this (Claude pro subscription does have this feature but it clones repos in its own env and you have to give it access to your GitHub account ) .. this custom solution works if you own a server which is reachable from your phone.. it’s not multiagent system .. it’s far simpler only for a lightweight AI coding assistant from the comfort of your phone.

#codeclaw
#codingassistant

[Chetan Yewale @3B0dyProblem]

@yajnadevam Could you explain more about this? I don’t see Krittika mentioned.

[Chetan Yewale @3B0dyProblem]

Genealogical adjectives ≠ ethnic or dynastic identity
•Terms like Mānuṣa, Aila, Nāhuṣa, Ānava in the Ṛgveda are descriptive patronymics, not proof of a permanent “Nāga umbrella identity.”
•Vedic hymns use multiple overlapping identifiers simultaneously, not rigid hierarchies.
•Nāhuṣa ≠ Nāga
•Nahusha → Nāhuṣa is a linguistic patronymic (“descendant of Nahusha”).
•Nāga is never defined in the Vedas as a synonym for Nāhuṣa.
•No Vedic or Puranic text explicitly equates Nāhuṣa = Nāga.
•False hierarchy construction
•The proposed ladder
Kuru > Bharata > Puru > Pañca-Jana > Nāga
is not attested as a formal Vedic social hierarchy.
•Kuru/Bharata/Puru are political lineages, not “subsets within Nāgas.”
•Pañca-Jana are not described as Nāgas
•Yayāti’s sons (Yadu, Puru, Anu, Druhyu, Turvaśa) are consistently treated as independent janapadas, not elite sub-clans of a Nāga super-group.
•No text calls Yadu or Puru “Nāga” as a primary identifier.
•Nāga has an independent mythic genealogy
•Nagas are repeatedly traced to Kashyapa–Kadru, with figures like Vasuki and Takshaka.
•This lineage:
•Exists outside Lunar genealogy
•Appears earlier than Nahusha
•Is never merged with Aila descent in scripture
•Nāga–Suparṇa (Garuḍa) conflict is mythic, not dynastic history
•Garuḍa is cosmic and symbolic, not a tribal emblem of Ikṣvākus in the Vedas.
•The Ikṣvāku–Nāga identification is retrospective allegory, not explicit Ṛgvedic history.
•Mahābhārata distinguishes clearly
•The epic keeps Nagas separate as:
•Subterranean beings
•Semi-divine clans
•Occasional marital allies
•It never redefines all Aila/Lunar peoples as Nāgas.

[Chetan Yewale @3B0dyProblem]

Logical Fallacies

1. False Equivalence
- Created fictional “Bhawandar” film about Pulwama to compare with Dhurandhar
- Argues: “If my anti-government fiction is propaganda, so is Dhurandhar”
Flaw: Comparing his fabricated narrative to a film claiming real-event inspiration is not equivalent

2. Double Standard
- Previously praised “Gangs of Wasseypur” despite its violence
- Now criticizes Dhurandhar for similar violence
- When confronted, claimed “I was wrong; back then I didn’t understand”
-Flaw: Selective standards based on political alignment, not principles

3. Circular Reasoning
- Premise: Film is propaganda
- Evidence: It mixes real events with fiction
- Conclusion: Therefore it’s propaganda
- Flaw: Assumes what he’s trying to prove

4. Cherry-Picking
- Focuses only on real footage and audio clips used
- Ignores that film is explicitly a spy thriller, not documentary
- Flaw: Selectively emphasizes elements supporting his narrative

5. Ad Hominem
- Called director “a disgrace to the name of art”
- Accused actors of “prioritizing money and fame over social accountability”
- Flaw: Attacks character instead of addressing arguments

6. Hasty Generalization
- Takes one dialogue about “Indian’s biggest enemy is another Indian”
- Claims it’s “often directed at Muslims living in India”
- Flaw: Extrapolates entire film’s intent from single line

7. Inconsistent Standards
- Says Dhurandhar is more dangerous than “Pathaan or Tiger Zinda Hai” because it uses real events
- Many films blend reality and fiction without being labeled propaganda
- Flaw: No clear criteria for when inspiration becomes propaganda

8. Echo Chamber
- Restricted video comments to members only
- Flaw: Avoiding opposing viewpoints and criticism

[Chetan Yewale @3B0dyProblem]

I had a thought - if the top tier companies were considered humans , what would their zodiac sign be.. and the AI gave the expected answer per the qualities associated with these zodiac signs. I guessed correctly for Apple, Google and Amazon.

Apple - Taurus
Luxury-loving, aesthetically obsessed, stubborn about their ecosystem. Values quality and beauty above all else, refuses to compromise on design principles.

Google - Gemini
Launches a thousand projects, abandons half of them. Brilliant but scattered, always chasing the next interesting idea. Masters of information and communication.

Amazon - Capricorn (I knew it..Amazon embodies the Capricorn qualities)
Relentlessly ambitious, long-term strategic thinking, workaholic energy. Willing to sacrifice short-term comfort for eventual domination. Everything is metrics and results.

Microsoft - Virgo
Detail-oriented, enterprise-focused, pragmatic. Went through a midlife crisis, came back healthier and more organized under Nadella. Loves systems, processes, and productivity.

Meta - Libra
Obsessed with connections and relationships, wants everyone to like them despite constant controversy. Always trying to reinvent their image and find balance between growth and ethics.

Tesla - Aries
Impulsive, aggressive, pioneering. Charges ahead without looking back. Thrives on competition and being first. Sometimes crashes spectacularly but always gets back up.

Netflix - Scorpio
Intense, transformative, knows all your secrets (viewing habits). Killed their old self (DVDs) to be reborn. Masters of psychological manipulation through content.

Nvidia - Leo
Currently the star of the show, loves the spotlight, confident to the point of arrogance. Everyone wants their attention (GPUs). Natural leader in their domain.

[Chetan Yewale @3B0dyProblem]

He could have explained and engaged more. But I won’t take that as rude reply as some others are implying.

Check his LinkedIn profile. He has been working in AI since the start of his career. So he certainly has the required capability of leading a team. https://t.co/P4Q30aq0UM

Another very plausible reason: maybe he does not want to disclose a lot of details on social media. Companies have policies on what kind of detail should be shared on social media.