The Project Died, I Didn’t.
A Chaotic Catch-Up on Life, Work, and Hacking Attention
Hey, Long time no see?
Says the one who hasn’t written a blog in half a year after writing a blog celebrating its one year anniversary.
It’s been 6 months since I updated anyone of you, if I write each and everything that happened then it would be a REALLY long blog, I don’t know if anyone would want that (even I don’t wanna type that much tbh).
Let’s just give a recap of the recent events (more like things which are in my memory).
Hemingway AI
Remember I was making an AI Scripting tool for marketers, so that they can just upload their current scripts and get another similar sounding script in like 10 minutes. We worked a lot on that. But for some reason you can say we lost momentum, everyone got busy in their own lives, one went to travel, another one was busy finding a job and I think I got lazy for a bit and by the time I got back some groove, it was too late (never say never). Still won’t say we are closing this off.
We serendipitously met some investors today and one of them urged us to push it to non-Indian audience for validation as it’s always easier to sell to anyone but Indians.
I think after half a year of inadvertently putting a pause on this, we are going to put a last stretch and market it to the right people.
Professional Life
Work is going well, can’t comment on the status of the company though, have been working on a really interesting project, you might have heard of platforms like bolt.new and lovable.dev etc. All these fancy prompt to website generators which one-shot development and have been taking the internet by storm.
After this there are platforms like rork.com and a0.dev and tempo which do the same thing but for mobile apps.
A little context about my company Apptile - we are a no-code drag and drop platform for making mobile apps, mainly targeted Shopify markets and got customers to use our app to create mobile apps after they have generated a good bunch of money on their website.
So coming back to where I was working on an interesting project. So we are making a competitor to rork and a0. A platform where you give a prompt and it converts it into a beautiful mobile app.
Our USP is that we have very good visual editing capabilities which are better than the current systems. Launch is coming soon.
Encountered a lot of things. How do platforms handle token management in each request does not cross the token threshold using mechanisms like Abstract Syntax Trees, Rolling Window Summarization, studying how GPT Engineer, Cline worked. Integrating a coding environment inside your browser.
Personal Life
Trip to Odisha
Attended a CAS concert which eventually got cancelled, still partied though.
Reddit, TikTok post, participated in a windsurf hackathon where I was having a mobile apps phase, basically that phase came because I wanted to promote apps on TikTok, I have been kinda intrigued about this niche of marketing your apps on TikTok.
Have been mildly regular to the gym and have been waking up at the same time for quite a bit now, sleep was pretty haphazard earlier
I didn’t really know how to ride a 2 wheeler before but I have been casually riding my friend’s scooty now, also a small win.
TikTok
Let’s talk about TikTok. TikTok is banned in India so no one even thinks of doing anything related to TikTok distribution but I was pretty interested to explore this paradigm.
I wanted to create apps so that I can market them on TikTok and see how the game works, for this I needed an iPhone, read in a lot of places, iOS content is boosted better and a deactivated US T-Mobile SIM, also read that having a US SIM in the phone tricks it to think you are in the US. So I did all that stuff and fortunately found out that my phone thinks I am in US, also had to buy VPN.
After that I needed a product, so I was going through Twitter/X and found a book reading product which seemed pretty marketable so I copied that, made a waitlist page at bookbopp.com and then integrated Umami analytics.
I posted around 6 random TikToks and 3 of them reached >= 1000 views, with one crossing 2K views. Doing this made me realize it’s pretty easy to get views on TikTok. Now my main focus here was to get some signups for my product but then that front didn’t work out very well, maybe cause my videos were not really optimized for that. Then I went on a thinking spiral if I really want some signups for my app or just wanna figure out TikTok virality and I figured out it was the latter.
I will experiment with it more.
I was pretty active on documenting my progress on Reddit also this time.
June 2025
Let’s come to the present time now. The initial team of Star Labs (Tushar, Sudip) with whom I created Hemingway.ink, we met after a long time, discussed deeply what were the issues which were responsible for Hemingway for not going the way we wanted. Those mainly revolved around alignment, motivation, prioritization.
This happened yesterday at a cafe where we serendipitously met a bunch of investors, sad part is we didn’t really have a product to pitch this time, but it was fun understanding their POV. A CEO of a wealth management fund told how he’s helping a person who made hybrid vegetables get connected to a VC and then how he’s acquiring another startup and how their Founder’s Office is making a product with 40 people which helps other family funds smartly see their assets and liabilities in one place.
We as a team are still brainstorming the idea, but we don’t wanna get stuck at an idea loop, so we are also making a product as of now for hiring. This is for a hackathon, we won this hackathon by the way last year.
It’s funny that my last blog had this hackathon and my new blog after six months also contains this.
That’s a lot of info.
Ciao.
PS – have been using a tool called Boardy AI, which calls me and I can talk to it regarding career advice and then it scrapes my LinkedIn and connects me to relevant people. Try it out.
