personal-update on S Anand

Things I Learned - 19 Jul 2026

Sun, 19 Jul 2026 00:00:00 +0000

This week, I learned:

Writing is slightly, but only slightly, better than typing (for adult learning.) One factor is that typing is faster, so many people take notes verbatim, summarizing and thinking less. ChatGPT + Claude
Graphology for personality is pseudoscience. ChatGPT + Claude
When I decide to spend time, or someone says “Let’s do X”, it’s worth checking: is this something AI can easily try, and is it clear to verify? If so, reinforcement learning loops could make AI good at it, making it a depreciating asset.
Studying how to live in an AI world is exhausting. (Not as bad as my MBA days, but not as easy as my data scientist days, either.) It requires me to make a larger mental shift, i.e. change my perspective, than I have since 2000, and that feels like work.
Both nl FILE and cat -n FILE add line numbers to files, but nl skips blank lines by default, cat doesn’t.
After using rtk for 2 months, I’m slightly downgrading it. It saves tokens but agents mess up shell commands when using it. It’s still probably a net saving, so I’ve changed my AGENTS.md from “Always prefix with rtk” to “Prefix supported, high-output commands with rtk… skip for bash builtins, pipes, loops, etc.”
I find 🔴🟡🟢 convenient status indicators in my notes. Similar ones are: 🟥🟨🟩, ❤️💛💚, 📕📙📗. I’m not fully convinced by: 😄😐😞, █ ▒ ░, ↑ → ↓, ▁▂▃▄▅▆▇, ■ ⬔ □, ● ◐ ○, ⚫ ⚪ 🔘, 🌕 🌗 🌑, etc. though they might have their uses.
Model updates means a SKILL.md and a plugin review / update, e.g. with GPT 5.6 Sol. So, like with any open source repo, use from people who update it regularly and benchmark it and version control it by model.
I asked Gemini 3.5 Flash thinking: “Which of our employees have worked on Microsoft PowerApps? Search @Google Drive and @Gmail”. It found one employee and a referral in under a minute. I asked ChatGPT with GPT 5.6 Sol with gws access. It found 3 more, plus 5 possibilities, in 12 minutes. Truly a rottweiler.
Parallel Search Turbo seems like a pretty good search API, especially for agents. Low price, high speed, and maybe good quality. #ForNow ChatGPT
Group chats in ChatGPT will probably get deprecated #ForNow.
What I learned from benchmarking my Ideation Protocol skill extensively:
1. Once you know the rubric, models can easily create a good prompt to optimize for a known rubric #ForNow. So rubric design matters more.
2. ⭐ Rubric design is really knowing what you want/need. To do this, iterating on output matters.
3. Position bias is real #ForNow. Always check if an (P, Q) comparison matches a (Q, P) comparison.
4. Models are still biased towards longer content, and potentially towards their own output #ForNow.
5. How to optimize a prompt or skill: Research and figure out what you really want, first. Then, ask a smart model for a prompt that optimizes for it. Benchmark only if you’ll use it a lot - it’s still a lot of work, and meta-prompting does a good job #ForNow. gbrain skillopt might be premature optimization.
You can use GPT 5.6 Sol in Claude Code #ForNow. (But what’s the point? Harnesses seem to be working better with their own models #ForNow.)
Our clients keep saying “We need to build a data lake” or “We need an enterprise data strategy.” I keep telling them, “No, agents can do it for you.” What I missed is: technology is the smaller part of the problem. Finding who has what data, getting access to it, and sorting out permissions (“governance”) is the bigger part.
Giving agents expert task-specific, testable procedures seems better than expert roles or mental models #ForNow. But benchmark in any case. ChatGPT
Python 3.3 introduced str.casefold(). It performs more comprehensive Unicode caseless matching than lower(); 'Straẞe'.casefold() becomes 'strasse'. (🟢 Unicode case-folding is standardized.) contextlib.closing(x) calls x.close() when its context exits. (⚪) In a dataclass, use x: list = dataclasses.field(default_factory=list), not a mutable literal default. (⚪) I learnt these while reviewing Codex-generated Python—illustrating, rather than proving, that reviewing AI-generated code can teach and catch errors. (🟡 Review remains useful across tooling. Review 2029.) “Do not discriminate against intelligence—artificial or otherwise” is a rhetorical value judgment, not an empirical conclusion. (⚫ Rhetorical value judgment, not testable. Review now.)
Here’s a nice idea from ChatGPT. “When itching to correct or clarify, FIRST restate their position to their satisfaction. ‘Did I get you right, fully?’” This emerged from the prompt suffix: Based on your research, and my past conversations, what are the top areas where and how (specifically) I can apply this principle on myself and others to maximize impact?
Automated evals can catch stuff humans miss. And vice versa. And given how many evals we create, we need automated evals to be written in an easy-to-review way. Do Automated Evals Work?
The BINEVAL paper reiterates that a bunch of Yes/No binary questions beats scales or ratings for many benchmarks. You know exactly how to grade and WHY you got a certain score. This is more reproducible and easier to learn from / act on.
When asked “How long will this software take?” models typically provide estimates assuming human speed #ForNow. Maybe they haven’t been trained enough on agentic timelines. So, when my colleague got a 2-4 week estimate which he was able to solve in hours, it was a surprise. (But, of course, it’s best to verify before promising speed.)
SKILL.md dramatically lowers the cost of learning a skill (since you don’t learn it - the agent does). That means that the value of creating skills is much higher - hundreds can use what you create (giving you recognition, if not money). I think I’ve underestimated the number of skills people will have available (I thought dozens - but it may be thousands #ForNow) and the number of skills people will create (I thought tens of thousands - but it may be millions #ForNow.) A Wikipedia (community curated, verified, high quality catalog) of skills might emerge #ForNow, if it hasn’t already.
Tacit knowledge is often just un-measured knowledge. Once I put a sensor on the bellboy’s hands at The Curzon Court, AI can figure out how he opens the door with the key and why I can’t do the same. The subset of tacit knowledge that’s AI-resistant is where attempts are expensive (“How to negotiate a merger” rather than “How to open a door”) and feedback is slow/vague (“Does the client trust me” rather than “Did the door open”).
The fact that Composio has ~20,000 tools is a market signal that connectors are commoditizing, and are a depreciating asset #ForNow.
A weak model needs a forgiving harness - which ends up slowing down model learning. Stricter, accurate verification environments are better for fastest model learning.
ChatGPT Work lets you run for longer, faster, install plugins and skills, host a website, etc #ForNow. It’s somewhere between Chat and Codex. It consumes Codex limits - something to watch for (since chat limits are quite generous).
Codex temporarily removed the 5-hour usage limit. Tibo. So, since I have 3 banked rate-limit resets #ForNow, I can, in theory, use 4 full weeks of Codex usage at one go. Reality: I don’t have problems large enough for a SINGLE week’s consumption!
From what I see of the State of AI Design and State of Prototyping, Figma is way ahead of competition #ForNow, e.g. Adobe, with Figma Make and Weave. I was also surprised how popular Cursor is (#2 behind Claude Code #ForNow). It’s also interesting that designers are coding directly #ForNow, using Figma just for edits / steering. But many research tools (note takers, survey analysis/research, etc.) will likely get eaten up by AI coding agents #ForNow, given how much designers are building their own tools.

My food preferences

Fri, 24 Apr 2026 14:09:36 -0400

I use ChatGPT to recommend which restaurant I should eat at and what food I should eat. So often that I decided to share a profile of my eating preferences.

But rather than think about it and type it myself, I asked it to

Efficiently interview me to identify my food preferences. Document it for AI agents to help me pick restaurants. Plan like an expert.

(Knowing ChatGPT, I also had add “efficiently” - otherwise it would give me a huge list of questions! Which it did that anyway…)

That makes it easy. Now, I just have to answer questions about my preferences, e.g. my budget range, restaurant styles, cuisines etc.

I also gave it a dump of my daily food log. (This year I’ve been logging all the food that I eat - for no reason actually - along with which of those I liked and disliked.) This proved to be a good idea, because it picked up something I didn’t realize:

You are unusually tolerant of repeating humble comfort foods at home, …

Very true. I ate curd rice day and night for 22 years at every meal. I eat Kaya Peanut Toast every day at Ya Kun Kaya Toast. (They make it the instant they see me standing at the back of the line, and miss me when I travel.)

So, here’s my tweaked version of ChatGPT’s recommended food spec on how agents should pick for me:

Eggetarian. No meat or seafood; eggs & milk are OK.
Prioritize novel cuisines & known exceptional dishes.
Prefer casual, functional, non-fine dining. SGD $5-20 feels fine.
Any cuisine works: Indian, Italian, Mediterranean, Middle Eastern, ...
Creamy > crunchy > gooey > chewy > airy.
Spicy is good. Heavy, raw, or loud are bad.
Desserts are great! Too sweet is bad.
E.g. creative fusion, chaat, inventive small plates, nutty flavors, hot cookies, textured desserts.
Prefer 10 min travel, 5 min wait. For exceptional food in a new city, an hour is OK.

PS: This is styled quite differently from how ChatGPT writes, but the content was correct.

Things I Learned - 29 Mar 2026

Sun, 29 Mar 2026 00:00:00 +0000

This week, I learned:

The Kids Should See This - great collection of videos for curious people. Thej
A jury fined Meta and YouTube $4.2m and $1.8m for building addictive features in their products. That’s a first. NY Times
“I think AI-type tools will actually revolutionize the experimental side of math, where you don’t care so much about individual problems and the process of solving them, but you want to gather large-scale data about what things work and what things don’t.” Terence Tao
The hedonic treadmill (which roughly quantifies a Buddhist principle) says that we revert to a happiness set point (which varies by individual). Worse, those who experience a high kick (e.g. a lottery) don’t get enough kick from normal wins (contrast effect) – Interactive explainer. The happiness neutral
As of today, a LinkedIn search for “llm psychologist” lists 9 people. I’m not alone!
- Anand S, LLM Psychologist, Singapore, Singapore
- Anshul Saxena, PhD, AI Advisor & Trainer | Technology Strategist | LLM Psychologist | Currently teaching humans, machines & business to work smarter through Generative AI and Quantum Computing | 15+ Years Experience, Pune, Maharashtra, India
- Charitarth (Chad) Sindhu, LLM Psychologist / Fractional Business & AI Workflow Consultant/ Digital Nomad, Tokyo, Japan
- Lancelot Salavert, LLM Psychologist, Barcelona, Catalonia, Spain
- Lior Dor(Durahly), Team Lead | Bug Banisher | Ex 8200, Tel Aviv District, Israel. Past: R&D Team Lead and LLM Psychologist at Superwise | A Blattner Tech Company
- maxime bodereau, Lead Creative Art Director | UX Forensics | Ai LLM Psychologist | Visual Alchemist | Codesmith | Brandologist | Full Stack Designer, Nantes, Pays de la Loire, France
- Mei Chen 🦋, LLM Psychologist | Lead Product Engineer | Delivering Agentic Experiences, Toronto, Ontario, Canada
- Shoshannah Tekofsky, LLM Psychologist at AI Digest, Zwolle, Overijssel, Netherlands
- LinkedIn Member, LLM, psychologist, mediator, Prague, Czechia
OpenAI acquired Astral!. This will likely slow down the new wonderful tools accelerating the Python ecosystem. Like with PromptFoo and OpenClaw, this seems to be about talent. The “acqui-hire” mode seems a clear niche career path now, and an alternative to getting hired (you get a much higher salary) or getting acquired (you take on much higher risk).
quickjs-emscripten lets you run isolated JS code securely in the browser, CloudFlare workers, NodeJS, and Deno. It compiles to WASM. @sebastianwessel/quickjs is a higher-level TS wrapper. Simon Willison
Manyana is a CRDT based version control system. It sounds like a good idea but I’m sceptical because merge conflicts are a “what should I do” problem more than “how”. With agents doing more merge conflict management, I am not sure this will offer a concrete benefit - but probably no harm either.
LLMs are able post-train LLMs on new topics. They’re improving fast. Jack Clark
Vibe Coding Fixer and AI Slop Cleaner are real job descriptions - which are morphing into enterprise offerings. But I still seem to be the only official LLM Psychologist
Notes from AI Services - Wrong Mental Models, Right Moment:
- AI services has 3 markets. Automatable work: vanishes in 2 years. Human-in-the-loop work: sustains. Judgement-driven: grows in importance.
- YC: don’t sell access to a tool for $50 a month, use the AI yourself and sell the finished work for $5,000.
- Sell output. Price on outcome. Sell to business, not IT.
- Sell accountability: proven success, with your guarantee.
- Sell authenticity: a brand story representing uniqueness, character, … or whatever… something people respect.
Data transfer between GPU and memory is a bottleck and three approaches are emerging. #
- Taalas is etching LLMs into the chip. Llama 8b runs at 17,000 tok/s (H200 is at 230 tok/s).
- d-Matrix is moving compute into SRAM memory chips. 30,000 tok/s for Llama 70b. Cerebras and MatX are similar: memory-oriented.
- FuriosaAI minimizes data movement. Groq and Sambanova are similar.
- But in the long run, commodity technology usually beats integrated stacks.
GPT 5.4 Nano ($0.2/MTok) and Mini ($0.75/MTok) are good options for bulk OCR, transcription, etc. as cost and quality comparable alternatives to Gemini Flash Lite and Gemini Flash. They can describe 75K photos for $50. Both models are better than GPT-5 Mini on most benchmarks.
Cool AI coding agent git prompt fragments:
- Use git bisect to find when this bug was introduced: …
- Find and recover my code that does …
- Sort out this git mess for me.
- Rewrite history removing …
- Split the last commit into multiple commits grouped logically.
- Start a new repo at … and build just this module … based on … with a similar commit history copying the author and commit dates.
Campaigns Are Knowledge Workers and the Tools Just Caught Up. A powerful framing. I saw this in action a few days ago when a friend was able to automate an outbound campaign with Claude Code.
EARS (Easy Approach to Requirements Syntax) is a simple structure for requirements. For example, “Users should be able to drag tasks between columns. The app needs to work offline too. Handle errors gracefully.” becomes the following - which AI can convert to and is easier to spot errors in. State machines and decision tables are useful alternatives, too.
- REQ-001 (Event): When the user drags a task card to a different column, the system shall update the task status to match the destination column.
- REQ-002 (State): While the application is offline, the system shall store task updates in local storage.
- REQ-003 (Event): When the application reconnects, the system shall synchronize locally stored updates with the server.
- REQ-004 (Unwanted): If synchronization conflicts occur, then the system shall display a resolution dialog to the user.
As of now, avoid using Claude.ai to create (large) visualizations. It runs forever and exhausts credits without generating anything. Claude Code works much better for this.

Vacation 2006

Thu, 16 Feb 2006 12:00:00 +0000

I am on vacation, and probably won’t update until 20th March 2006.

Comments

Madhu 20 Feb 2006 10:04 am: Happy Holidays:)
Sai 22 Feb 2006 4:58 pm: The new colour looks better!
Krishna Shirali 8 Mar 2006 10:10 am: Have a relaxed time
PP 14 Mar 2006 10:48 am: Simpler; better.
ravi 21 Mar 2006 2:30 pm: looking forward to hear ur holiday escapades

Vacation

Tue, 19 Apr 2005 12:00:00 +0000

I am on vacation.

Comments

sathish 20 Apr 2005 3:39 am: enjoy madi.. have fun.
harish 21 Apr 2005 10:57 pm: hi anand. have fun
mark wu 22 Apr 2005 2:31 pm: yes and thank you
Venkat 25 Apr 2005 6:01 am: Enjoy maadi
Dhar 25 Apr 2005 12:36 pm: Coming back here after a long break. Have quite a few interesting links to check out! Will certainly check out the MIT Prank once I am on a good Net connection.
sunshine 29 Apr 2005 8:46 am: Anand u r so cute he he he
sunshine 29 Apr 2005 8:47 am: Hey that one was to pull ur leg(i read that u hate being called cute"
bharani 29 Apr 2005 10:26 pm: this commenting style doesnt look very clean…leaves a lot of blank space on your page…i was wondering if there was some pic there that was not showing up…lol…
S Anand 1 May 2005 8:54 am: True, Bharani. I’ve been thinking the same. Am planning some changes, and will have them online by mid-May.
sathish 4 May 2005 4:08 am: Anand, One suggestion is that you could have the comments below the main post, instead of right hand side of the post.
rinku 9 May 2005 4:59 am: rinku:where did u get a copy of “What do you care what other people think"by Richard Feynmann.I have been hunting for that book for around 2 years and still no success.
Venkat 16 May 2005 4:55 am: Where are you right now Boss?
sp 16 May 2005 8:53 am: hi “rinku”…i have a old copy of the book, and you
Sriram 23 May 2005 6:21 am: I work for Dell and they seem to have blocked your weblog!! Any clues why?
Dhar 25 May 2005 10:20 am: Or if you dont have a server outside that you can use, try searching on Google for something called JAP (Java Anonymity and Proxy)

Unwell

Mon, 05 Apr 2004 12:00:00 +0000

I was unwell. I’m almost back to normal now, and likely to update a little more frequently.

No more updates for a while

Fri, 01 Nov 2002 12:00:00 +0000

I’m going to try and not update until Monday (11th Nov, not 4th). Don’t even bother visiting. Happy Diwali!

I am on vacation

Mon, 27 May 2002 12:00:00 +0000

I am on vacation. Please don’t expect updates till 20th June 2002. (Sorry!)

Web is boring and life is interesting

Fri, 10 May 2002 12:00:00 +0000

No, I haven’t stopped updating because I wanted my Google entry to be on top for a long time. It’s just that the web is boring, and my life has become interesting otherwise. Read more in Mumbai. I’m in Delhi these days, so I hope to add a Delhi chapter as well.

On vacation

Thu, 27 Dec 2001 12:00:00 +0000

I’m on vacation till 2nd Jan.

I am at Bangkok

Sun, 24 Jun 2001 12:00:00 +0000

I’m in Bangkok. Not planning to update this site for a week. Incidentally, it’s almost time for my site’s first anniversary – a year since 28 July 2000, when I started updating it regularly.