linkedin on S Anand

What I don't post on LinkedIn

Sat, 06 Jun 2026 11:55:10 +0530

I don’t post all my writing on LinkedIn. For example:

Fewer strategy posts, e.g. “Where Enterprise AI is Headed”, “How My Innovation Team Works”, etc. aren’t on LinkedIn.
Fewer developer posts, e.g. my AGENTS.md, my SKILL.md files files, CLI tools, evals, etc. aren’t on LinkedIn.

I also shorten content because of LinkedIn constraints. For example:

No links, e.g. the list of all my AI-in-education resources
Short content, e.g. my full advice for teams using AI is much longer than the LinkedIn post.
Trimmed prompts, e.g. how to convert meeting transcripts into a personalized org-consulting report
Snipped chats, e.g. the full moves of GPT-5.5 playing chess

I filter the LinkedIn posts, sharing what’s most useful for most people.

But some people asked if I can share the full content over email.

So, I’ve set up my full blog, by email (free, via Google Groups): https://groups.google.com/g/s-anand

Preview before you subscribe. Switch to weekly digest if it’s too much.

Learnings from building Babbage Insight

Mon, 19 Jan 2026 18:50:19 +0530

Here’s a great post by Karthik Shashidhar on why they shut down Babbage Insight, and the learnings from the experience.

(I’m reproducing in full here since LinkedIn is hostile to content.)

I added ⭐ to points I found most interesting.

As the more perceptive of you would have figured out by now, we are shutting Babbage Insight . When I told this to one of my old friends, his immediate reaction was “so what were your learnings from this experience?”. And so I decided to write this.

Remember that I’ve written all of this primarily for my own benefit and future reference. Any cost / benefit from this to anyone else is only incidental.

This is also in no particular order. And the salt you need to add is that everything is purely with the benefit of full hindsight.

With full benefit of hindsight, this (AI for data analytics / insights) has been a rather tough sector to operate in. Nobody has really cracked it here (not even established players - distribution doesn’t imply traction), and there has been a bloodbath of startups, which (unfortunately) we are also part of

⭐ When you require access to a customer’s data warehouse, it doesn’t matter what size the customer is - every sale is an enterprise sale.

From our customers’ perspective (which we belatedly realised), this was a risky buy. They didn’t know what they wanted, and they couldn’t see how what we were offering would clearly help them.

⭐ Hiring is massively underrated, and a massively difficult job. We hired poorly. Maybe the biggest skill required for a startup founder is hiring.

Tenure match with your first set of investors. Raise early if you are raising. That way you don’t run out of energy before the money runs out (happened to me)

I got into the “where’s my idli” (or “where do i put mine” - both translate to the same phrase in Kannada) syndrome. Every time I read about something someone else in the sector was doing, I got defensive, wondering if someone else is already doing what we wanted to do; rather than figuring out how to learn from or copy the idea. Perversely, despite running a startup, I got into a zero sum mindset (with all my competitors), which didn’t help

I should have started with consulting, and then used learnings from that to build products.

Even if you are building a product, build it with a specific customer in mind. Building with a generic / hypothetical customer can mean that you build for no one

Our attempts to build a “generic enough” product were to no avail, as each customer wanted their own set of customisations. And we made the mistake of trying to tweak the entire product to fit these customisations, rather than hack.

We hear romantic stories of people who cracked big-figure cross border sales, but those are driven by spectacularness bias. You need to be in the market that you are selling to. There is no other way. If you cannot, you need to employ someone in that market.

I listened to too many people. I tried doing too many things that don’t come intuitively to me, and so struggled.

Every single one of my burnouts has come about when I’ve tried to write production grade code. I like coding but software engineering simply demands too much attention to detail for me to do well. And I keep making the same mistake - step in to write code when the team cannot. And burn myself out

I love talking to (potential) customers, but I don’t have the discipline for sales. Following up on time. Following through on intros. Etc.

⭐ Hire a small number of highly skilled, even if expensive, people. At early stage startups, especially in the age of AI, people who just implement what you tell them to do are overrated. LLMs imply that you don’t have 10X engineers any more - you have 100X (apologies for that cheesy line). And you need to hire a few of these, else you’ll be stuck.

I absolutely hated putting my location as “United States” on LinkedIn, but had to do it for sales purposes. Did a lot of other such things, because they are “done things”.

Poor hiring meant I spent too much time writing code, and too little time building the business.

I never got a coherent communication of what exactly we are building. Obviously this was a huge problem

Our product was doomed the moment I wrote the first line of code.

We didn’t use LLMs enough. Our code was too deterministic. This was a function of our hiring.

We didn’t vibe code enough.

We didn’t move quickly enough. Again it was possibly a function of our hiring.

I’m scheduling this post to be sent at a time when I’m in the middle of my Vipassana. I’m pretty sure I’ll return and double the length of this list!

LinkedIn is hostile to content

Mon, 19 Jan 2026 18:39:33 +0530

It’s incredible how hostile LinkedIn is for reading / writing content.

Posts containing links to external websites (like my blog) get significantly less reach. That’s why you see links in comments, not the post!
You can’t copy content from posts on their mobile app. You can’t even easily select the entire article on the web app! Selecting a part, and then shift-clicking elsewhere (which works almost everywhere) doesn’t work.
Also, the copied text isn’t clean. It’s filled with hidden text (e.g. “Skip to search”), duplicated text (e.g. author name repeated), and other junk.
It’s hard to export content. For example, the export feature does not include the original links in your articles, nor the links to images you posted!
It’s hard to scrape content. LinkedIn actively tries to prevent scraping, and their TOS prohibits it.
No formatting. You have to embed unicode characters.
Search is terrible. You can’t search for posts by keyword, date, or author easily.
No public posting - so you need to log in to read anything.

The only reason I’m on LinkedIn is because of the network. My current approach is:

Write on my blog: https://www.s-anand.net/
Copy paste on LinkedIn, formatted
Link to the original, and to hell with the reach

Over time, … let’s see.

Things I Learned - 26 Oct 2025

Sun, 26 Oct 2025 00:00:00 +0000

This week, I learned:

Before founding a place to do good, work in a place that does good and learn. Ben Werdmuller
What should we teach when vibe coding becomes good enough for non-coders? Ethan Mollick
- Problem decomposition
- Clear communication & spec writing
- Core technical foundations: file systems, access control, networking, APIs, version control, data structures, databases, deployment
- Software development skills: Debugging, Testing, Refactoring, Design patterns, UI/UX
- Project management: requirements, prioritization, scoping, …
Codex CLI tips:
- codex --add-dir $DIR lets you write into $DIR
- codex --full-auto is the equivalent of codex --sandbox workspace-write --ask-for-approval on-request
Terse code is not necessarily easier or harder for LLMs to write. It’s about how unusual (or not aligned with training data) the code is. Gabi Teoduru
How are people using browser agents like Comet / Atlas? Simon Willison
- Most popular: YouTube video summaries with timestamps
- Most useful: Form filling: Government forms, data entry, repetitive bureaucratic tasks
  - Foreign language navigation: Applying for pension in Korea, navigating sites in other languages
  - Time reporting auto-completion
  - Insurance claims: Reading policy documents and drafting appeals (successfully got claim reimbursed in India)
  - Compliance training click throughs
- Next most useful: Shopping / planning
  - Energy provider comparison - Comet checked current plan vs competitors on Check24, calculated exact annual savings per provider
  - Financial tracking: Finding Amazon orders, tracking Airbnb spending with refund calculations, analyzing bank transactions
  - Trip planning: Mapping 50-100 places on Google Maps automatically
- Interesting: Airport shuttle discovery - Found shuttle that user missed in manual searching
HubFS mounts GitHub repos on the file system. Every file system action directly works on GitHub via a REST API. Useful for some scenarios but less useful for note-taking than something like GitDoc which offers a delayed sync.
Ernest Ryu solved an open problem in convex optimization using ChatGPT. Quotes:
- ChatGPT is now at the level of solving some math research questions, but you do need an expert guiding it.
- ChatGPT was really effective at accelerating my progress. This work took about 12 hours, spread over 3 days. In hindsight, the proof is really simple.
- But I iterated through so many other strategies that didn’t pan out, and ChatGPT crucially helped to quickly explore and eliminate those dead-end approaches. Also, the key successful steps were suggested by ChatGPT.
- ChatGPT did not produce the proof in a single prompt. The process was highly interactive. It generated many arguments, roughly 80% of which were incorrect.
- Yet some were genuinely novel to me. Whenever I recognized a novel idea, whether correct or only partially so, I distilled the key insight and prompted ChatGPT to develop it further.
- My contribution:
  - Filtering out incorrect arguments and accumulating a set of correct facts.
  - Identifying promising new lines of reasoning and guiding ChatGPT to explore them further
  - Recognizing when a strategy had been fully explored and deciding when to move on.
- ChatGPT’s contribution:
  - Producing the final proof argument.
  - Significantly accelerating my (or our) exploration of the many dead-end arguments, rapidly ruling out approaches that did not work.
Comparing the GPT 4.1 and 5 models at all different of reasoning, I’ve switched my default from GPT 4.1 mini to GPT 5 mini (medium). Far smarter for a slightly higher cost. Artificial Analysis
python -m pdb -c continue script.py or uv run -m pdb -c continue script.py runs a script and drops into pdb on unhandled exceptions (post-mortem). ChatGPT
Technology removes constraints. We then do what we really value. Claude
- When writing became digitized, we stopped cared about spelling/handwriting for its own sake. Spelling bees and handwriting classes declined. “ur” is acceptable.
- When fitness tracking became easy, many just track, few exercise more. Few people value exercise
- When GPS became ubiquitous, we stopped learning geography. Most value arriving, not knowing
- When photography became unlimited, most captured moments. Few perfected shots
I had Codex scrape my ~2,000 pending invites on LinkedIn and asked ChatGPT to analyze it. Here are learnings: ChatGPT, private
- Power-law. 5% of inviters account for ~42% of all common connections. Top 10 people alone for ~20%.
- IITM student invites are high (~14%), but with 0-2 common connects, i.e. distant strangers.
- EdTech is tiny in count but has the highest common connections per person (outlier-sensitive but real).
- Among ≥20-commons, many hold VP/Head/Site-Lead titles in Data/AI or GenAI (not just recruiters).
- GenAI people are 7-8% and steady across months. Not a useful signal to prioritize.
- Premium ~ Senior. Premium accounts show ~40% senior titles vs ~29% for non-premium.
- Finance invites have higher seniority rate and more common connects than healthcare.
- Followers have higher common connections (~6 vs ~4).
⭐ Memory can be code. Agent memory is anything it choose to persist. Agents can write code on the fly to automate tasks, save them, and serve the code on the next request, potentially modifying the code as required. This is like the conscious mind saving a habit for the subconscious to execute fast.
Finally: Microsoft Office has an agent mode that lets you talk to it and do stuff. The Verge

Things I Learned - 06 Jul 2025

Sun, 06 Jul 2025 00:00:00 +0000

This week, I learned:

When adding a coding benchmark for LLMs, here’s a question I’d like to add. #benchmark

How do I use Apache Arrow in the browser via cdn.jsdelivr.net to create a .parquet file and download it? Give me minimal working code I can paste in the browser console to test.
LinkedIn has an undocumented link that shows schedules posts at https://www.linkedin.com/share/management/ which redirects to https://www.linkedin.com/feed/?shareActive=true&view=management

Here’s a JS snippet you can paste in the DevTools console of an npm package version page (example) to get a Markdown list showing the versions and dates

copy(
  $$('table[aria-labelledby="version-history"] tbody tr')
    .map((tr) => {
      const a = tr.querySelector("a");
      const date = new Date(tr.querySelector("time").getAttribute("datetime")).toLocaleDateString("en-GB", {
        day: "numeric",
        month: "short",
        year: "numeric",
      });
      return `- [${a.textContent.trim()}](https://npmjs.com${a.getAttribute("href")}): ${date}.`;
    })
    .join("\n"),
);

DuckDB can read JSON APIs! Ref
⭐ When bringing in humans-in-the-loop, applications must make it easier to review and to edit the work.

Sun, 20 Oct 2024 07:32:39 +0000

Arun Tangirala and I webinared on “AI in Education” yesterday."

“(“Webinared” is not a word. But “verbing weirds language”.)”

Mid-way, Jose Swan from the audience asked, “Can you summarise this session using an AI?

There are SEVERAL tools you can use to summarize talks. Whisper for transcription, FFMpeg for keyframe extraction, #NotebookLM for podcast generation, text-embedding-3-small for topic modelling, and of course, any regular LLM include #ChatGPT for summarization or translation.

I spent this morning applying those to the webinar video, and writing it up the process.

Read on for the details…

https://www.s-anand.net/blog/tools-to-publish-annotated-talks-from-videos/

PS: I referenced Raj Vadigepalli midway and thought of Vivekananda Vedula, bringing back hostel memories from Indian Institute of Technology, Madras.

Mon, 22 Jul 2024 03:19:27 +0000

Loved this Rocky Aur Rani Kii Prem Kahaani scene where Ranveer asks, “Chinese ko Chinese bol sakte hai?”

हम बहनदी भी नहीं बोल सकते? आंटी, मैं दिल्ली से हूँ। मैं कैसे नहीं बहनदी बोलूं बहनदी!? कैसा जमाना आ गया है? फैट-ों को फैट नहीं बोल सकते, ब्लैक-ों को ब्लैक नहीं बोल सकते, ओल्ड-ों को ओल्ड नहीं बोल सकते, मुँह खोलने से डर लगता है मुझे! आप मुझे बताओ, चाइनीज़ को चाइनीज़ बोल सकते हैं?

Can’t we even curse “Damn it!”? Aunty, I am from Delhi. How can I not say, “Damn it”, damn it!? What times are upon us? You can’t call fat people fat, you can’t call black people black, You can’t call old people old, I am actually afraid to open my mouth! Tell me, can you call Chinese people Chinese?

अगर मुझे चाइनीज़ आर्डर देना है तो क्या करूँ? फ़ोन करूँ “हाँ भाईसाहब, वह हिमाचल से राइट जो जगह है, हाँ हाँ पड़ोस वाला मुल्क, हाँ वही कोरोना वाला, हाँ, वहां का एक क्रिस्पी चिकन लगा दो” अब आप बोलोगे “कोरोना वाला” नहीं बोल सकते!

If I want to order Chinese, what should I do? Should I call, “Yes bro, that place to the right of Himachal Pradesh. Yes, yes, the neighboring country, yes, the one with Corona, yes. Get me a crispy chicken from there” Next, you’ll say, “You can’t say “Corona”‘!

बचपन से, बचपन से दादी कहती आ रही है, “चाय पीने से काले हो जाते हैं, चाय पीने से काले हो जाते हैं”। अब पता चला है, ये रेसिस्ट है!

Since childhood, Grandma taught us, “Drinking tea turns you black. Drinking tea turns you black”. Now I learn that’s racist!

किसी ने ये भी नहीं सिखाया कि कौन सी गाली देने से मिसोजनी हो जाती है। यह भी नहीं बताया कि गोलू को गोलू बोलने से फैट शेमिंग हो जाती है। हमको तो यह सब नॉर्मल लगता। हमको तो कभी रॉन्ग लगा ही नहीं। हमको पता ही नहीं यह बोलने से सामने वाले की फीलिंग्स हर्ट हो सकती हैं!

No one ever taught us which abuses become a “misogyny”. No one ever told us that calling Chubby ‘Chubby’ is “fat-shaming”. We all thought this was normal. We never had an inkling it was wrong. I didn’t know that saying this may hurt others’ feelings!

Beautifully worded. Lovely acting. Nothing I could add to it.

(Except: Whisper and most transcription APIs handle Hindi poorly. Claude 3.5 Sonnet does better than GPT-4o on Hindi grammar.)

Winning the alphabetical race

Mon, 20 Nov 2023 06:48:58 +0000

Since my name (Anand) begins with “A”, I used to get called on fairly early at school. In attendance. Answering questions. Classroom exercises. Quizzes. Even the distribution of test results.

A few people later told me that it is good training, since I’d always be prepared. (Maybe. I’ve no idea.)

At IBM and IIMB, Ajit was the only one ahead of me, alphabetically. Then he went a step ahead and named his son Aadi. I thought that’s impossible to beat.

Today, we recruited Aabhas Bharadwaj. I checked on LinkedIn. I can’t find a single name on LinkedIn that’s ahead of his, alphabetically.

So, does he win the alphabetical race? Can you find one ahead of his?

Comments

RK 22 Nov 2023 8:58 pm: :-) Interesting observation! I know two women from 2 different states of India who are named Aabha
Jaspreet 8 Mar 2024 8:58 pm: Yes, Aabha was one of my friends in school! But, never thought of such knd of race…… interesting!

Sun, 01 Jan 2023 23:48:45 +0000

And THIS is impact. Thanks for making MY 2022 an amazing year, Tanvi Bansal!

Thu, 25 Aug 2022 23:00:07 +0000

I can’t phrase this better than Sunil Sharma who said “As if the struggle in life is not enough, they put the Oreos so beneath th Marie Golds :-(”

So true. Shame on those who ate the Oreos on top and didn’t dig into the bottom, Gramener team! Work hard next time 🙂

“PS: Has anyone observed something like the “Oreos at the bottom” phenomenon in your office?

Mon, 06 Jun 2022 16:19:37 +0000

I was surprised to see a big chunk of my #music library is from the 1960s – thanks to the songs my parents grew up with, and passed on to me.

In celebration of Music Day on 21 Jun, I’m treating myself to 2 weeks of songs from the #1960s.

I’ve linked to my collection. Hope you discover some gems too!

Wed, 04 May 2022 09:45:08 +0000

When Sunil Kardam said Gramener is the one place he got to apply all his skills - management, engineering, people skills and analytics - it reminded me how cross disciplinary Gramener is.

There’s something to be said for generalists in the era of deep focus.

Wed, 16 Mar 2022 09:48:22 +0000

We inaugurated a new office in #Hyderabad.

10 years ago, I wrote on my blog: “I would create a small company that would probably fold. Then I’d do it again. And again, 10 times, because 1 in 10 companies survive. And finally, I’d be running a small business that’d be called successful by virtue of having survived. A modest, achievable ambition that I had the courage for.

Naveen Gattu, Mayank Kapur, Ganes Kesari, J. Ramachandran, Ravinder Ireddy, Vengatesh CR. – we’ve come a long way 😃

Gramener successively crossed 10 employees, 10 clients, 10 years, $10 mn and is on its way to 10 offices.

I have the same request as 10 years ago.

It’s scary but exciting. Please wish me luck!

Mon, 28 Feb 2022 05:10:55 +0000

I’m planning to publish a 3-hour self-paced #onlinecourse.

But I don’t know which topic would be more useful.

One topic is #datascience tools for non-programmers.

Another is a step-by-step guide to #datastorytelling for analysts.

What’s more useful for you?

Could you share with people, so I work on the more useful course? (Thanks 🙏)

Thu, 30 Dec 2021 07:03:53 +0000

I read 52 #books in 2021.

The best #nonfiction was The Almanack of Naval Ravikant. Succinct & deep. I can spend a decade practicing every sentence.

The best #fiction was Brandon Sanderson’s Rhythm of War. A brilliantly rich magic system, and what a plot, what an ending!

My one-line #reviews of the books are below.

http://www.s-anand.net/blog/books-in-2021/

Tue, 21 Dec 2021 13:39:09 +0000

Can foreigners break into #Hollywood?

A break into films is hard. Particularly when you’re a foreigner. But is Hollywood more open or less open than other countries?

If we go by iconic actors, the numbers are not encouraging. Just 2% of Will Smith’s co-stars are non-Hollywood. But over 30% of Jackie Chan’s co-stars are non-Chinese.

But #clustering shows that Hollywood is actually among the most open to foreigners. #dataanalysis

Wed, 13 Oct 2021 08:19:06 +0000

Arvind Venkatadri put together a #datavisualization course in #R titled “R for Artists and Designers”."

Beautifully crafted. Brilliantly structured. I loved it!

https://r-for-artists.netlify.app/

Tue, 05 Oct 2021 06:27:54 +0000

Over 30 people have created data comic stories for the #DataComicgen awards.

Getting insights from data is hard. Telling stories from these is harder. Telling comic stories is the hardest.

Yet, these two dozen stories simplify data into simple (and even interactive) comic narratives.

Hats off, participants. Hats off!

https://gramener.com/comicgenfriday/awards/gallery2021/

Sun, 15 Aug 2021 06:19:49 +0000

The most under-rated #geospatial capability is spatial joins, IMHO. It lets you draw correlations from #geodata.

During the 2019 assembly elections in India, it allowed me to join voter data (by constituency) with religious population (by district) and show that BJP’s vote share increased with Muslim population, while it reduced the AIMIM-allied JD(S) vote share.

That’s just the start. We can explore:

Where to open new schools?
Where to locate bank branches?
Does wealth lead to more or less theft?
Does poor health lead to more or less pharmacies?
… and hundreds of other random questions.

My day job doesn’t let me explore these much. I’d love to work with someone, though.

Would anyone like to explore such correlations? Which one(s)? I’d love to support you.

Talk video: https://youtu.be/ql3Kz8ZjhRg

Transcript: http://www.s-anand.net/blog/maps-delimitation-and-gerrymandering/

Wed, 28 Jul 2021 03:03:35 +0000

I’m really looking forward to this Data Comicgen #event.

Get the #data on 5 Aug
Analyze it with #googlesheets
Use Comicgen for #comics #storytelling
Submit on 26 Aug

It’s a great opportunity to find fellow data storytellers and comic enthusiasts – to see their work and share yours. And win awards.