š©š½āš»Product @meet_dandy - Building the Digital Dental Lab | Prev Data @ Apple, Amazon Healthcare | @nyuniversity | Musings on Product, Data & AI
Joined October 2024
- Tweets101
- Following577
- Followers33
- Likes2K
If you currently work in Data Science (and this does not include MLEs, DEs), I would push you to consider what your role as a Data Scientist (embedded in Product, Business, Finance, Marketing, or whichever domain) looks like in your current company in the next 6 months.
If you are content with the answer, then great. If not, the next question I would ask is how can you evolve your scope to also own the business outcome and not just the delivery of insights
Building the sleep apnea program at Dandy is quickly teaching me that building a fully vertically integrated business is not for the faint of heart
YES! So excited for this era of software.
My own version of this is once you anchor on what problem you are solving and why it matters - that clarity helps me context switch between shipping a feature, standing up a sales motion or running a marketing campaign.
I saw this tweet from Anne right when I got back from a three week trip in Southern India and I knew I had to come back to it. This was essentially the theme of my trip home - fears around AI.
Across family, friends and strangers, ranging from a 13 year old to my 95 year old grandmother, every conversation came back to two types of questions: 1)Will AI replace me, 2)How will my business survive?
With each of these conversations I tried to use a consistent approach - I would hear them out, ask questions about their concerns, learn about what a regular day looks like for them, and slowly nudge them towards thinking about the possibilities of AI.
This was not easy, and I don't think I got anywhere close to where I wanted to. But I learnt two things:
1. You really have to meet people where they are. My friend's dad runs a small business in Bangalore. What worked was asking him what his day to day looked like and trying to find where an AI assistant could help him
2. A lot of people don't even see the problem. A common question I got is "Why would you even want to do that, what are you saving?". What this group needs is an AHA moment, where we have solved a core problem for them. And this is a product and marketing problem, not a model problem.
So what would an ideal version of this look like? India runs a lot of small businesses on @WhatsApp . Now imagine @Muse proactively surfacing and completing tasks for a small business owner, meeting them on a surface they already know. How amazing would that be? This likely is already on the roadmap, and this is the version of AI I am most excited about.
We can all benefit so much from AI, but we need to care as deeply about the messaging. If you are working on getting AI into the hands of people outside of tech, I'd love to hear what's working.
This week, someone I met only once a year ago, called me out of the blue to ask me what the heck is going on with AI.
Today, the woman who runs the farmers market I go to asked me if AI was going to kill us all.
Iāve never received so many questions about AI from people outside the industry ever since I started learning about AI in 2017. And none of the questions are about products, getting help, or upskilling. Itās fear and concern.
There are so many ways AI can help people, and sure, this could be a segway to more meaningful conversations about that with non tech industry people. But right now so much of the general public is reacting to it like itās a mega death laser.
Maybe this narrative is working for the frontier labs in the ways they care about. But Iām really disappointed that so much of the narrative has gone in this direction. People are scared and they donāt even understand what happened, why, or what it means.
Not enough people tell you that finding the right partner can significantly improve the quality of your life.
And I donāt mean right in the sense of a framework or a checklist - whatās right for me is unique to me versus what might be right for you.
I have been programmed to hustle, itās taken years of reflection to learn to slow down and Iām still learning. But having a supportive partner has helped compound that growth - like earlier today when he stopped me out of the blue and said āDo you realize how amazing it is that youāve built these relationships over 20 years?ā And I walked away thinking, damn, I actually did that. How wonderful is it to have someone that sees something in you that even you donāt recognize!
Feels good to be back home in Bangalore!
Any recommendations for tech events happening around the city? Would love to learn more about building in Bangalore!
Outside of that, first stop is Umesh Dosa Point. Nothing hits like a Benne Dosa from Umesh!
What are the odds that I would read this paragraph on the same day as this tweetš
One thing I have never enjoyed is being defined by a single job description/title.
Even as a kid, I did not want to be only an engineer, or only a data scientist or only an artist - I would seek ways to blend my interests.
In my current role I get to sit at the intersection of data, product and commercial. I get to take an idea from identifying a customer pain point to research to building. AI plays such a huge role in making this possible for one person. What a wild time to be in tech!
And the blending doesn't stop at work. Next up, I am exploring the intersection of art and engineering. More on that soon
One thing I have never enjoyed is being defined by a single job description/title.
Even as a kid, I did not want to be only an engineer, or only a data scientist or only an artist - I would seek ways to blend my interests.
In my current role I get to sit at the intersection of data, product and commercial. I get to take an idea from identifying a customer pain point to research to building. AI plays such a huge role in making this possible for one person. What a wild time to be in tech!
And the blending doesn't stop at work. Next up, I am exploring the intersection of art and engineering. More on that soon
Revathi Anilkumar retweeted
The main reason I love @shreyas is because the way he thinks and what he writes travels seamlessly beyond product.
This is rare in any field.
There are many great product thoughts out there. But ironically most thoughts will help you at product while *keeping you in product* instead of enabling you to think bigger than product and above product. While Shreyas helps you with product, the content is doing more, making you suitable for leading or becoming a founder.
Most fields are rife with ideas and thinkers who help but imprison practitioners. The best advice is the stuff that helps and sets you completely free.
Just thought this is worth calling out. It has come up in several of my coaching sessions and my clients werenāt as consciously aware of how important it is to exercise caution with the advice that seems sound but puts them in a box.
A friend and I were talking about how much traveling has changed for us with AI. You can instantly look up information on the go and learn about places.
Last week I was in Montreal, and I noticed that there were a lot more heritage buildings converted into newer spaces. Felt like the ideal balance between old and new. But this also felt unique to Montreal.
So I asked Claude, and learnt that most North American cities, a building is classified as heritage if someone opts in, however in Quebec the province can just designate buildings. And the city is now required to survey old buildings and list potential heritage buildings.
Finally got to reading this gem by Shreyas and the following quote stood out the most to me.
This quote also perfectly summarizes the most recent season of The Bear!
Claude Chat for those who want to understand this idea more deeply. This one Claude Chat contains more practical insight for ambitious & talented product leaders than entire MBA programs and most business books, biographies, and profiles.
claude.ai/share/64129ddd-ec5ā¦
Watching the latest season of The Bear reminded me of Slackās notification system - there is so much effort that goes into getting even the simplest of things right.
It blows my mind of quickly things can move at smaller companies. You can just put out an intention to do more certain type of work, begin to lay the foundations for that work, and within 10 days, become the person responsible for it
Revathi Anilkumar retweeted
Good to remember your product is not growing because (1) it doesn't solve a burning pain or (2) the product experience isn't good enough that people tell their friends.
[Tweeted as I'm inventorying some of the things I work on into these two brutal buckets, and wanted to share.]
This has been a pretty big unlock for me over the last few weeks. When Iām looking to brainstorm, I start my prompt with voice mode enabled and āThis is a brain dump, ask me follow up questions and help me structure my thoughtsā. Has worked well so far
One pattern I find useful for working with LLMs is a nice long ramble session. Sometimes the LLM needs more bits to understand what you're trying to achieve, but you're too lazy to type them. In these cases I like to lean back, switch to /voice and just ramble for like 10 minutes, total mess, anything goes, full stream of consciousness. Sometimes I declare it up top, something like "switching to speech recognition sorry for any typos...". Sometimes I turn it into a small interview of a few turns. But I find that the LLMs are somehow very good at reconstructing long incoherent rambles and often their echo of your own tangle of thoughts comes out quite a bit cleaner than what you started with. The result is that you improve the mind meld and have to correct things less from that point on.
I am always in awe of parents that can talk to their kids as friends.
Last week on the subway, I was sitting next to a father and son (~5 year old). The father was line drawing objects and people, taking his son along the journey while drawing. He was drawing this old school picnic basket that this woman was carrying and heās gently nudge the child through the process, with questions like - so where do you think sheās going today, what do you think sheās carrying, do you see how round the basket is. And the kid would follow along, responding each time.
What a beautiful relationship!
Here for Bayesian stats, but also here to appreciate the aesthetics of Asheās work - line drawings, a few instrumental horns. Thinking of the importance of craft in helping ideas stick!
