Meet the Writer: Two-Time Founder Sam Bhattacharyya on Accidentally Finding Product-Market Fit

Let’s start! Tell us a bit about yourself. For example, name, profession, and personal interests.

My name is Sam. I grew up in the US and have a background in old-school AI/ML from Columbia and MIT. After grad school, I started my first startup (an e-learning app in West Africa), which we pivoted 10 different times [1][2][3] until we got acquired by a customer during the pandemic.

I worked as the head of AI for the acquirer (a company called Hopin in 2024, when it was acquired by a private equity firm. I then started my 2nd startup, Katana (AI Video Editing), it’s going ~~super well~~, ~~terribly~~, and unexpectedly

Profession-wise, I’m not sure I neatly fit into any clean category. I definitely have an engineering/CS/technical background, I’m definitely a founder, and on paper I’m CEO, but really I’ve spent my time, both in my startups and when working for Hopin, doing a bit of everything (writing code, talking to customers, training AI models, marketing, HR/Management).

I don’t necessarily do all of those things well. I find it easier to prototype an idea than to build a stable production app. I find content marketing and talking to customers easier to do than b2b sales. If there’s one thing I’m particularly good at, it’s highly experimental tech ideas (like, I invented/patented a new video compression algorithm), and more recently, I’ve found a niche in developing really efficient AI models, especially those that run in the browser and those that deal with video/computer vision.

Like, I’m not a good AI researcher, but I can look at a real-world user problem, decide that none of the established research/open-source stuff is adequate, build custom architectures/training strategies for that specific user problem, and deploy the results into a working product with customers, and that approach has been the core of the last 3 tech products I built (AI Filters SDK, Free video Upscaling tool, AI Video Editing app).

On the personal front, I’m Indian-American, and I currently live in Mexico with my wife (who is Mexican) in a town called Querétaro. I have a 6-year-old son who really likes Minecraft. If I have one hobby, it’d be learning languages - I speak Spanish and Mandarin fluently, and actually speak to my son in Mandarin.

Interesting! What was your latest Hackernoon Top Story about?

My latest HackerNoon top story, How a Demo Page for my Abandoned Open Source SDK Accidentally Found Product Market Fit, was about how my 2nd startup is going - my first product (an AI Video Editing app) is going okay. Still, recently it’s been overshadowed by a random side project (or more specifically, the demo page for an open source AI Upscaling SDK) which unexpectedly got product market fit by itself, and which now has ~100,000 Monthly Active Users, which is 200x the traffic that my “actual” startup product has.

Do you usually write on similar topics? If not, what do you usually write about?

I have written quite a bit in the past, mostly about my last startup. I primarily used to write on Medium (you can see my past articles here) and I just wrote them organically, some of it was about specific projects we were working on, some of it was just my own re-counting of the startup journey, and the given the fact that we launched out of MIT, moved between Accra Ghana, Lagos Nigeria, Bangalore India, Querétaro Mexico, Boston, New York and San Francisco, pivoted 10 times, nearly went bankrupt 3 times, got a video compression patent, unexpectedly found product market fit and were acquired by Europe’s fastest growing startup all within 5 years, I feel like we genuinely had some interesting stories to tell.

It was only after this article, about how our AI Filters SDK was 10x more performant than Google’s own open source models, that we got inbound organic customer interest.

I often also have been writing update emails to my first startup’s investors, and now my current startup’s investors, although those are emails, the style, tone, and length are similar to the article I just wrote.

I do want to write more stuff going forward, mostly about the stuff I’m working on. My current adventure with a random open source up-scaling tool has motivated me to re-consider a few more interesting open source projects I’ve wanted to work on (an open source version of my first startup, and I swear, if you can believe me, I believe I have developed an algorithm than can improve the speed & accuracy of transcription models by an order of magnitude).

Those are things which I might write both academic papers for, but also developer-friendly explanations of how they work and why they matter.

I’ve also worked on some random side projects in the past, which have paired well with an article I wrote - like I feel like this article I wrote a few years ago could have been of interest to hackernoon readers. For me, the primary goal isn’t the writing in and of itself, it’s more - work on genuinely interesting things, and writing is a vehicle to tell the world about it.

Great! What is your usual writing routine like (if you have one?)

I don’t really have a writing ‘routine’ as a defined ‘process’ that I stick to, though maybe I should if I do more of it.

For almost every article I’ve written, there’s been some kind of story to tell, and I usually just write from start to finish in a ‘train of thought’ manner.

This tends to mirror my work style in general. I like it when I work intensely, deeply focused on one task until it’s done. This works just as well for writing articles as it does for writing code. Just like with writing code, yeah, I don’t do JIRA, I just wing it.

After years of writing investor emails, pitch decks, and blog posts, I’ve instinctively just internalized how to structure stories and narratives - I don’t really plan or storyboard it; if there’s any planning, I just write some bullet points down (the headers) and just start writing. I add images and graphics that I think make sense as I write it.

I hardly do any editing afterwards, though more recently I’ve added spell checking as a formal step because I find myself writing so fast that I just make typos left, right, and center.

That’s also how you’ll know none of my stuff is AI-generated 😅, I have too many typos for anyone to think my articles were generated by an LLM.

Being a writer in tech can be a challenge. It’s not often our main role, but an addition to another one. What is the biggest challenge you have when it comes to writing?

I don’t really have any complaints about writing. I think the hardest part is just budgeting time to do it, because it isn’t my main role, it’s in service to whatever it is I’m working on, but honestly, for me, it’s part of the job description anyway.

Like, sure, debugging code (especially AI-generated code) takes time, but it has to be done, and it just takes time.

What is the next thing you hope to achieve in your career?

I want a project, other than this upscaling tool that randomly found product market fit, to work out. Like, again, going back to the transcription thing I was talking about, if I really could build something that could advance the state of the art of an industry by an order of magnitude, if the world’s largest companies used my algorithm/software, that’s a career-defining kind of thing.

Wow, that’s admirable. Now, something more casual: What is your guilty pleasure of choice?

I miss my childhood. Adulting is hard, and for a while, I had dreamed of curling up in my bed and playing Pokémon like I did when I was 8.

At some point, I was sick last year during the holidays while visiting my parents’ house. I pulled out the same old Gameboy I used when I was eight, and I tried playing Pokémon Red, and… I couldn’t. Too much time has passed, my brain has changed, and I just don’t have the patience to play a video game where it’s like “Go on a quest to do xyz”, like my real life is full of to-do lists. I don’t really have the patience to chase down 150 Pokémon when I’m already frustrated with chasing down the 10 items my wife asked me to pick up from the grocery store.

My only solace was listening to old video game music, which was all the nostalgia without any of the effort. Whoever made a lo-fi version of the original pokemon sound track, I want to truly and deeply thank you.

Do you have a non-tech-related hobby? If yes, what is it?

Like I mentioned, learning languages. I’m fluent in Mandarin and Spanish, and I speak to my son in Mandarin. I have tried to pick up others (I have some level of proficiency in Portuguese, French, and Italian, but I always mix them up). I have tried learning Japanese and Arabic, but it’s just where’s the time?

I at some point realized that if I don’t constantly maintain my Mandarin, it will decay, and I’d rather be fluent in one language than be barely conversational in 5. To that end, my little free time watching movies/videos is all in Mandarin.

Honestly, I watch tons of kid shows (Peppa Pig, Blippi) in Chinese. I used to watch those shows with my son, but now I watch them by myself

(1) Because he’s too old for them now, and

(2) because those shows have by far the most practical vocabulary. Like, I can read the New York Times in Chinese, and I used to, but when am I going to talk to my son about the Ukraine-Russia conflict or the Trump presidency?

I do need to tell him to wash his hands, though, and kids’ shows like Peppa Pig and Blippi are the only places where you’ll regularly encounter normal, real-world household vocabulary like “Water faucet” and “Bicycle pedal.”

I have a 6-year-old son, so I don’t really have time for my own hobbies; any free time is for his hobbies. He really likes playing Minecraft, though, so if playing video games counts as legitimate 21st-century Father/Son bonding, that’s a win for me.

What can the Hacker Noon community expect to read from you next?

I have some ideas for articles. Like for one, I was going to eventually write an open, insider’s perspective on the rise and fall of Hopin (formerly Europe’s fastest growing startup)

I hope one of the other projects also works out, and if they do, you’ll certainly hear from me. I may also port (and update) some of my older interesting side projects here.

What’s your opinion on HackerNoon as a platform for writers?

I am surprised by how democratic this is. Most social media platforms have algorithms that can make it feel ‘random’ as to what posts and content do well and what doesn’t, especially when you don’t already have a big network/audience.

For someone without a big pre-established audience, I find it genuinely refreshing and hopeful that there is a platform that both has incredible reach and readership, and provides a platform for good content, even if it’s from writers who don’t have a big established network or audience themselves.

It motivates me to actually write more, genuinely interesting stuff - previously, the barrier was always “I can write this interesting thing but it’ll just go into the void”, and having a platform where good content can actually bubble up to a real audience is appreciated for someone just getting started.

Thanks for taking the time to join our “Meet the writer” series. It was a pleasure. Do you have any closing words?

See you on the next post!