Mary Meeker didn’t just chronicle the rise of the internet — she helped the tech world understand its velocity. Her Internet Trends decks from the ‘90s became required reading in boardrooms and pitch rooms alike. Since then, through Kleiner Perkins and now BOND Capital, she’s mapped how technology reshapes behavior, markets, and power. Her reports don’t just reflect momentum — they pre-empt it.

Her latest: Trends — Artificial Intelligence (May 2025) is a 340-page document that unpacks how AI is not just a product wave, but an infrastructure shift. It goes deep: 800 million weekly users on ChatGPT, $212 billion in annual CapEx from Big Tech, and a clear inflection where training costs are ballooning while inference becomes cheap and scalable. It’s a shift in the stack — not just in tools, but in who can build, deploy, and scale.

The most important parts aren’t flashy. They’re foundational. The report examines how open-source models are quietly gaining ground, how the cost dynamics are redefining compute strategy, and why the next dominant AI players might not be U.S.-born. It’s not written for tourists. It’s written for builders.

If you're a founder, you owe yourself the time. This is more than a trends deck — it’s a calibration tool. It shows you what’s happening under the hood of the biggest platform shift since mobile.

📥 Download the full report (PDF)

Bengaluru is one of those cities where startup momentum feels real — you just need to know where to look.

This guide by Plum makes that easy. It brings together the most relevant communities, accelerators, investors, and voices that help you get started, find your people, and plug into what’s happening.

📖 Read the full guide

If you're new to Bangalore or want to explore the ecosystem with more intention, this is a great place to begin.

Some standout insights:

Why I keep sharing this:

It’s not hype. It’s not a scene. It’s just genuinely useful for any founder trying to find momentum in Bangalore.

I end up recommending this to almost every founder who’s moved to the city or is here for a few weeks. It saves time, opens doors, and makes the city feel a little more yours.

👉 Read it here

Before they became well-known, these products were just trying to get 100 people to care.

This timeless piece from 2020 by Lenny Rachitsky explores how leading consumer apps like Reddit, Figma, Duolingo, and Clubhouse found their very first users. It features 40+ real examples of early traction that came from hands-on effort, not hype.

📖 Read the full piece

It’s still one of the best reads for founders figuring out how to launch and grow something new. The tactics may vary, but the mindset is the same.

Some standout insights:

• Reddit created fake user accounts to simulate engagement in the early days.
• Figma ran in-person design workshops and gathered live feedback from students.
• Duolingo quietly spread through niche forums and communities before any major campaigns.
• Clubhouse focused on bringing in the right early users instead of chasing big numbers.

Why this matters now:

Even with all the new tools and channels, early growth still depends on real conversations, community, and effort.

This read is a good reset for anyone thinking about growth. It brings the focus back to building trust, doing the work, and being close to your earliest users.

If you're planning your own 0 to 1, this one is worth saving.

Google’s not dying — but it’s no longer the only gatekeeper to discovery.

In this sharp and timely piece from a16z, Zach Cohen and Seema Amble introduce GEO (Generative Engine Optimization) — a new way to think about how people find answers through tools like ChatGPT, Perplexity, and Claude.

📖 Read the full piece

For founders used to chasing SEO rankings and keyword hacks, this shift is massive. GEO reframes the game: your startup won’t win by being on page 1 of Google — you’ll win by being the cited answer inside an AI output.

Some standout insights:

Why this matters now:

 The LLM layer is becoming the new frontend of the internet. Whether you’re building a B2B product, a consumer app, or a media brand — your visibility increasingly depends on how these engines perceive and reference your work.

This read helped me realize: distribution isn’t just about reach anymore. It’s about relevance in the eyes of machines.

Curious to hear: how are you adapting your content, product, or growth strategy for the GEO era?

I started working on 58Miles in 2023 with a simple thought. Let’s make a better bag. Something I would actually want to use every day. A bag that works the way we live and move.

It’s now 2025. Still working on it. Still tweaking. Still not calling it “done.”

Everyone is launching apps and lots of AI apps, and I am launching a bag. Not because the world needs a new one, but because I couldn’t find a single one that worked for me.

Most bags either look good but don’t function, or they’re white-labeled, mass-produced pieces with designs that make no sense. Jahaa dekho, wahaa cool bags and good looking bags, they are like black holes. Too many zips. Random compartments. Loud branding. All show, no soul.

And pen compartments. Pen ke compartments me kaun pen rakhta hai.

I wanted something built for people like us. People who work on the move, who travel for work, who dump things into their bags and still want to stay sorted. I wasn’t trying to create a fashion statement. I just wanted a bag that works.

So I started making one. Here in India.

At first, I thought it would be quick. A few prototypes, a couple of months, and we’d have something good. But it never played out that way. Every sample fixed one problem and introduced another. The laptop section would feel too tight. The strap would lean weird. The bag wouldn’t sit right when placed on the floor. The zip didn’t feel smooth enough. These were small things. But they mattered.

40+ samples and 2 years, all these for little things no one notice but feel.

It’s hard to explain why this took so long. The truth is, I didn’t want to carry a bag that I had to explain. I wanted something that made sense the moment you used it. The kind of product that gets out of your way and becomes part of your rhythm.

I wanted to build for the people like us, who work, travel for work, build, create, hustle, messy and still organized.

Picture abhi baaki hai mere dost.

..

This post is part of Slice of Startup Life on eChai Startup Stream — where founders share the unfiltered version of their journey, as it’s happening.