Aswath Krishnan

I have fancied starting a business for a very long time.   Finally after getting my green card, building a financial safety net, learning by working at other businesses, and being inspired by the recent breakthroughs in AI, I took the plunge recently!  I'm sharing some reflections from my first few months.  Fair warning: all of these may change over time as I experience more and grow wiser.  1. Businesses are incredibly hard This is repeated often, but we still often underestimate or forget it as we are constantly inundated with victorious stories of very, very few survivors.  Finding and sustaining the right partners, ideas, customers, business models, teams, and investors is a long and tough journey. By default, your business is dead. You have to will it into existence and life every day with perseverance, smarts, support, and luck. There are many ups and downs, and tricky and hard decisions along the way.  I now empathize and understand why founders...

Find some quiet and relaxed setting to think through these questions.  1. What would you advise a friend? We are more clear-headed when we step back from the situation.  2. What feels more exciting and fulfilling? What would you do if you weren't afraid and if others won't know?  Fear and norms often hold us back.  3. What kind of life and lifestyle do you want?  What option ranks higher in your rubric? Having a structured approach with top factors and coarse ratings for each path+factor lets us be more objective and increases confidence in the decision.  4. What is the best-case scenario, most probable scenario, and worst-case scenario? Can you protect the downside? Outcomes are usually non-deterministic, so think in possibilities and probabilities.  5. Is this a one-way door or a two-way door? Most decisions are reversible. Spend more time and focus on those that aren't.  6. What are the 2nd and 3rd order effects? What happens after this decisio...

I distinctly remember this anecdote from Stephen Covey that I read many years ago.  At a workshop, a troubled participant approached him and confided that he and his wife weren't in love anymore. He desperately needed help to save his marriage.  Stephen Covey's advice was simple:  "Love is a verb. Love – the feeling – is the fruit of love, the verb or our loving actions. So love her. Sacrifice. Listen to her. Empathize. Appreciate. Affirm her." Covey's reframe is powerful. He changed love from something that just happens to people to something they can deliberately do.  I think the same advice applies to the universal goal and challenge of enjoying life. If you want your life to be more enjoyable, you must enjoy life.  Enjoy is a verb. It's a choice, a mindset, and a way of living, thinking, and acting.  In fact, the word enjoy originates from the French word "enjoier", which means "to give joy" or "to take delight in". It is a s...

Even the most ardent free-willers will have to confront the origins of their will at some point.  Have you ever wondered what shapes your mind and behaviors? The influences that nudge us towards success or failure, happiness or despair? What makes our life unfold in the way it does? Let me share a mysterious experience from my high school years that puzzled me for a long time and only started making sense to me years later. MAX  When I was in grade 10, I went to a large book fair with my family. Like all events in India, it was bustling with people and energy. In the midst of the chaos, my family and I were drawn towards the stage that was particularly abuzz. The stage banner displayed the name ‘MAX’, short for 'Madras Academy for Excellence'. A group of people, who were national memory record-holders no less, were captivating the audience with their ability to quickly memorize long lists of words or facts and answer any questions about them. I was impressed and intrigued beca...

As Product Managers or entrepreneurs, we often have multiple good ideas and an exciting vision. That's necessary but not sufficient.  It's easy to come up with a large collection of many capabilities and features that can together make a product compelling to users and produce business impact. But that isn't practical.  Most individuals, startups, or teams within larger companies get defunded or lose steam before they can do everything they envision if they don't produce results along the way. The complexity of more also confuses users and overwhelms teams.  A startup founder wisely told me, "Startups don't starve...they drown." What's really important is identifying and prioritizing the minimal set of ideas or bets that are going to have the largest impact, and then distilling them to their core essence . The 20% of work that'll produce 80% of the results. You need to have the conviction and discipline to say no to everything else. Only if you do ...

Traditional software simply executes explicit instructions provided by Human Intelligence. This blog is powered by software. And like most software we use, it functions based on explicit instructions. When you are on a blog post, the code explicitly specifies to deliver content from a CMS or database that's tied to the blog post and then styles it as per explicit HTML and CSS-based instructions. When you are on the home page, the code explicitly asks to display the list of 50 recent blog posts from the database.  This software is really useful as it can execute more precisely, quickly, cheaply, and on a larger scale than humans. But it is also simply doing what the human developer explicitly instructed it to do.  Enter - Software powered by Artificial Intelligence. Human instructions not needed.  Let's take this ChatGPT Plugins demo.  ChatGPT is able to:  Create a meal plan Generate a visual of the meal Shop for the ingredients Tweet about it.  No explicit ...

🤯 1)  I had my first "Aha! moment" with #ChatGPT / #LLMs in November 2022, just around when ChatGPT was released.I had asked ChatGPT to draft the annual strategy for the company I was working at, without providing any additional context. I had only used ChatGPT to answer basic questions or generate poetry and didn't expect much in response to something more complex. But the answer that ChatGPT produced in a few seconds resembled what the exec team had drafted, in content, reasoning, and priority. A few more prompts and minutes later, ChatGPT also had answers for how to present the plan, organize the team, mitigate risks, and execute the plan. ChatGPT was able to replicate a lot of the strategic work of the highest-paid and most experienced people at the company in a fraction of time and cost! 🤯 2)  I experienced my second Aha! moment when I restarted programming last month after a hiatus. After some trial and error, I'm now able to get ChatGPT to now generate 75%-...

We have been talking about the progress of GPT (or ChatGPT) across versions 1 through 4. We measure them by the number of parameters, test scores, capability, range, accuracy, modalities, plugins, etc. We celebrate and are excited about the progress of every version. There are so many memes now hyping up what GPT-5 would be.  It just struck me that humans, as individuals and as a collective system, are also a form of intelligence. And humans have also been upgrading versions. We are on Human2023 right now. Wouldn't it be fascinating to similarly explore and celebrate the progress of humans and speculate about the future? Thanks to a technology developed by Human2022/23, it is now possible to generate a sweeping narrative that transcends the boundaries of time and space with a single prompt. Read on to take a step back, traverse the epochs of human evolution, celebrate the milestones of our past, and gaze with wonderment at the horizon of possibilities that awaits us.  ***...

I just watched Air, a fun and inspiring movie about how Nike, then an underdog in the shoe industry, managed to sign Michael Jordan, a rookie at that time. The movie starts with Sonny Vaccaro (played by Matt Damon) opposing his colleagues' decision to go with the low-risk and low-reward move of signing three average basketball players. Teams and companies operate a lot based on group discussions and consensus. Those don't usually lead to brilliance or extraordinary outcomes - they converge on average, which is why most companies slowly wither and die as they grow. But Sonny doesn't care for the consensus. His deep passion and expertise in basketball make him realize that Michael Jordan is going to be a superstar before most people did. He aims higher and dreams of signing Jordan, defying his bosses and other business restrictions. He makes bold moves, breaks rules, and risks his entire career to make it happen.  As this impossible idea starts to take shape and seem more pos...

Those who control the pipes of information, control the world.  It's a notion I have heard and generally agreed with. Humans are powerful agents in the world. Humans use information to act and human minds can be easily programmed by information. For instance, the "Stop the Steal" movement in the 2020 US Presidential Election, where disinformation led to a significant number of people believing the election was rigged. Controlling information is, in essence, controlling people. But it wasn't until a recent personal experience that the true impact of this control hit me. I was excited to launch a new business and decided to run some ads on Meta/Facebook. Imagine my surprise when, less than an hour after submitting my ad, it was flagged for a policy violation. I suspect an automated risk check program had detected some inexplicable issue and also penalized me for being a new ad publisher with a new page. Although the violation seemed thematically relevant to my ad, it wa...

I don't plan to fill my blog with AI-generated content but felt this was pretty interesting to share.  AI progress can accelerate exponentially if they help humans make it better. Then AI gets better and helps humans make it even better, and so on, until it can make itself better, eventually leading to singularity and superintelligence.  In this prompt, I asked it to take cues from biological intelligence to come up with ideas to improve AI. I'm realizing that it's a good prompt pattern to ask AI to combine influences from different fields to solve a problem, since that's how a lot of good ideas and inventions are generated, and that isn't easy for humans to do.  Prompt:  Imagine you are developing an intelligence. It's still nascent, but you want to accelerate its powers. what would we do? look at examples from evolution and biological intelligence development. for e.g. humans got some brain capabilities like imagination, and also opposable thumb and tools and ...

An age of unparalleled peace, joy, love, creativity, and discovery for all of humanity. Fueled by an abundance of wisdom, resources, science, technology, labor; and elimination of all suffering, friction, and bottlenecks.  Made possible by the invention of benevolent superintelligence. *** Picture this day in your life.   You wake up feeling like your best self - well-rested, lucid, energetic, and joyful. You are surrounded by people you want to be with, also in the same state of bliss.  What's on the agenda for today? Anything you want! You have no chores, commitments, or worries.  Surf on a tropical beach with friends and family, both alive and passed? Granted, you are there. Don't know how to surf? No problem. You are going to learn it from the best coach or use an assisted board.  Hungry? You are served delicious and nutritious dishes invented just for you.  Feeling unwell or tired? You get diagnosed and treated in no time.  Need some inspiration o...

If you are an analytical person like me, you think a lot. You are compulsively problem-solving, analyzing, plotting, ideating, imagining, what-if-ing, and worrying. You even pride yourself on being a thinker.  This is undeniably useful, sometimes. You may come up with good ideas that can make your life better. You may proactively anticipate problems and avoid them.   But compulsive thinking comes at a significant cost.  The cost of not being at peace or enjoying the present. The cost of moving through time and space without paying attention to the mystery and beauty around us. The cost of being a slave to the finicky, obsessive, and paranoid mind.  The cost of not living.  The thinking that's trying to help us survive is taking over and not letting us be alive .  Alan Watts said, "A person who thinks all the time has nothing to think about except thoughts. So, he loses touch with reality and lives in a world of illusions. I'm not saying that thinking i...

See Part 1 discussion on how to generate ideas here . This covers how to execute on ideas and how to take care of yourself Generated by Bing GPT chat, with no edits.  Part 2: How to execute on ideas This is Bing. I can generate a conversation between Paul Graham, Sam Altman, Garry Tan, Jessica Livingston, and Michael Seibel on how to work on startup ideas after you have identified 2 or 3. Here is an example of how they might talk: Paul: Hi everyone. Thanks for joining me today to share some insights on how to work on startup ideas after you have identified 2 or 3. I think this is a common challenge for many founders who are trying to find product-market fit. Jessica: Absolutely. I think it's important to remember that the very best startup ideas tend to have three things in common: they're something the founders themselves want, that they themselves can build, and that few others realize are worth doing⁴. Sam: That's right. And once you have a few ideas that meet those crit...

Generated by Bing GPT chat, with no edits.  This is a possible conversation between Paul Graham, Mark Zuckerberg, and Sam Altman where they discuss and build on each other's principles for coming up with startup ideas.  Paul: Hi Mark and Sam, thanks for joining me today. I thought it would be interesting to talk about how we approach startup ideas and what we've learned from our experiences. Mark: Sure, sounds good. I'm always curious to hear how other founders think about startups. Sam: Me too. I think there's a lot of value in sharing our perspectives and insights. Paul: Great. Well, let me start by saying that I think the way to get startup ideas is not to try to think of startup ideas. It's to look for problems, preferably problems you have yourself . Mark: I agree with that. That's how Facebook started. I wanted a way to connect with my friends at Harvard and see what they were up to. I didn't think of it as a startup idea at first. It was just somethin...

A popular friend of mine once shared a trick to liven up any gathering.  "I just make up a game." He went on to demonstrate by making up a game of throwing a crumpled piece of paper a wall and then making it land in a trash. The first to 5 wins. Three of us were then engrossed in the game of trash basketball.  Games make boring things fun. Think of any sport - cricket, soccer, football, tennis, car racing, etc. They are all pretty meaningless in a practical sense, but billions of people are engrossed in them.  That's a key idea because many important things in life, like exercising, filing taxes, learning something new, chores, working towards an idea or business, etc. have parts that are pretty mechanical and boring. They become barriers for us to doing them well and consistently, and eventually, stop us from succeeding at the important thing. You can try to use willpower or motivate yourself with the final goal, but it's tiring and often futile to resist boredom.  ...

Couple of years ago, I reached out to a smart person to see if he wanted to invest in a friend's startup.  His response was quick and confident.  "I don't invest in seed startups. It’s not a great asset class unless you have some systemic advantage like y combinator. But I appreciate your offer."  He is right. Most early stage startups fail. To win in early stage investing, you need to invest in large number of promising companies, so that a few of them can 100X and pay for all the rest. In order to do that, you need deal flow, enough capital, or a different way to pursue a portfolio approach.  The broader lesson is to first recognize when you are drawn into playing a different game in business or life, then step back and understand the nature of the game, what's needed to win that game, and decide whether you can and want to develop the skills and advantages to play and win the game.  

This is probably the most valuable question you can ask when working on anything. A very common failure mode for founders, PMs, software teams, and people in general is to try and solve poorly articulated and unsubstantiated problems. So they end up solving non-existent problems or solving them poorly.  I think this fatal error can be easily avoided with a simple framework - "Problem Stories" organized in an "Opportunity tree".   What is a Problem Story? User stories are a popular and useful tool to describe building blocks of a solution.  As a User X, I'd like Y, so that I can do Z They offer more clarity and context than just saying "build Y", and that reminds, empowers and motivates the team to build the right thing.  Problem story is exactly that, but for the problem.   As a [Person/User] A, I have [slight/significant] problem with B, because of C. Supporting by data X and/or anecdotes Y.  For example:  As a a person trying to lose w...

Four years ago, I spent $12 to register www.aswathkrishnan.com and then spent a few days on and off wrestling with Blogger to create a free but presentable blog. And here we are now - reading the second sentence of my 150th blog post!  Creating this place to write and developing the practice of writing are among the most significant things I have done for myself.  It has given me the space and the reason to reflect, think, consolidate, distill, and record my experiences, thoughts, and insights. It has been a consistent source of clarity, equanimity, and solitude in an increasingly noisy and complicated world. It has reached tens of thousands of readers, some of whom have reached out and shared satisfying notes of thanks and appreciation. It has also sparked interesting conversations and connections, and even helped me land jobs.  But above all, it has given me my own canvas.  Unlike the canvas at a job, this is a place that I can always call my own. A place where I h...

Most of us aren't fans of chores. So "favorite chore" sounds like an oxymoron and it's even surprising that billionaires do any chores at all.  So what is this chore and why?! Ta da!  It is dishwashing. Jeff Bezos joked in a 2014 interview, "I do the dishes every night. I'm pretty convinced it's the sexiest thing I do". Bill Gates similarly revealed on a Reddit AMA the same year, "I do the dishes every night." and further explained that it helps him clear his mind.  I also stumbled on the magic of dishwashing on my own. It has been the perfect chore for me in the morning or whenever I'm feeling sluggish for a few reasons.  (1) It's physical and gets me moving. But not too physical, like a workout, that my mind resists. Moving the body and doing physical things like scrubbing plates and touching warm water is one of the simplest and fastest ways to get more grounded, mindful, and energetic, especially in our increasingly sedentary an...

Languishing Naming a problem or an uncomfortable emotion is a good first step towards overcoming it. This is why there was collective relief and aha moment when Adam Grant diagnosed the strange feeling that everyone was going through during the pandemic as "Languishing" in his now famous NYTimes article .  Languishing, he says, is the neglected middle child of mental health. It is that middle void in the wide mental spectrum between depression and flourishing. It is that experience of life where you don't feel actively depressed, but you also don't feel excited or engaged about your day-to-day and your life. You are just passing time, going through the motions, and getting by without a sense of joy, fulfillment or purpose. You feel dispassionate, dull, and indifferent.  The acknowledgement of this experience and addition of this term to our lexicon might be one of the gifts of the pandemic. Because this isn't just a passing pandemic phenomena - it is a very longst...

Can something be simple but difficult? At first glance, that seems contradictory. But if you think on it, you'll realize many, many problems and goals are simple but difficult.   For example, being above-average healthy* is quite simple. You need to eat, exercise, and sleep well. The solution is well known, has only a few key dimensions, and can be explained and understood pretty simply. But it is also extremely difficult to get started on that, break your existing patterns, and do it consistently with delayed gratification for a long time.  Same can be said for developing good relationships, creating a blog or small business, saving money, being at peace, raising a child, etc.  A common mistake people make is conflating difficulty and complexity. For instance, instead of acknowledging that the difficulty in being healthy is in execution or consistency, people may think the fault is in the simplicity. Then they turn the simple solution into something complex, like ec...

I recently had an unpleasant experience on a deal with an organization that had an unscrupulous track record. In retrospect, I realize that's exactly what I should expect when doing a deal with the devil! I had entered the deal despite knowing and having an uneasy gut feeling about the devil's unsavory past. This is classic. Well-meaning people make deals with the devil because of a few possible reasons: (1) they don't know or assume good intent, (2) feel they don't have a choice, (3) they are tempted by the upside, (4) they think the devil's changed or they can transform the devil, (5) they think they can out-devil the devil. Almost always, they end up regretting making the deal.  Devils are cunning masters of psychology. They can appear grand, altruistic, powerful, and charming. They know how to make you feel special with sweet talks and gestures. They can character shift like chameleons. If cornered about their past, they defend, deflect, create moral ambiguity, ...

I am outraged! Aren’t you? They are out there Saying lies   That defy us! Wait, wait   Who are they?   Who is us? Weren’t we all Just one? And these lies   What are they? Do they hold No truth, frustration Or hidden pain?     Listen I didn’t listen But I do know. Their team and tone   Say enough. We must retaliate Or we’ll lose!   Maybe we can   Pause and truce. Mingle together And learn more.    We may fare better   In some middle ground. You silly   This is war! Haven’t you heard? Choose your side   Or step aside. Only one will remain They or us!

I went back to Chennai, my hometown, and visited my grandmother in December 2022. It had been a long four years since my previous visit and this was my wife, Daljit's, first time meeting my grandmother and extended family. We had planned this trip for the March of 2020, but had to be postpone because of the pandemic.  As we entered her room, her face lit up and so did mine. She greeted me fondly and then quickly chided me for losing weight and asked me to eat more and exercise well, instead of dieting. She complimented Daljit as "beuuutiful" multiple times and even conversed jovially in English. She asked me how long it's been since our marriage and then assessed that it's now time to have kids.  She was a few days shy of her 90th birthday and she seemed noticeably frailer compared to the last time, but in good spirits and lucid as always.  Then I asked her how she is doing.  She smiled, brought her hands to her chest, closed her eyes slightly, and replied, "...