Rohan

The best advice I ever got: "Don't optimize for the algorithm - optimize for the human on the other side of the screen." Social media success isn't about gaming the system. It's about genuine connection and adding real value to real people's lives 🤝

its the classic "do as i say not as i do" situation AI design tools are great for quick mockups and iteration but when it comes to brand identity - the thing that makes people feel something before they even understand your product - you still need a human who understands emotion and culture the irony is almost too perfect. "use our AI to replace your designer" says the company that just spent $500k on a branding agency lol

OpenAI hired Jony Ive. Gamma, Perplexity, and Superhuman all hired the same branding studio. Every AI company selling “you don’t need a designer” is quietly paying designers to build their own brand. The companies closest to the technology are the ones least willing to trust it with.

the compliance industry has been running on vibes and spreadsheets for way too long biggest problem i see with most startups is they treat SOC2 like a one-time exam instead of an ongoing practice. get the cert, forget about it for 11 months, then panic continuous monitoring changes the whole dynamic. you're not just "compliant" on paper, you actually know whats happening in your infra in real time

I started Secureframe because too much of compliance was theater. I believed cybersecurity and compliance automation should make organizations genuinely more secure, not just "more certified." That means real continuous monitoring of your cloud infrastructure and core business systems. Gaps surfaced in real time, not in spreadsheets reviewed once a quarter. And independent, AICPA peer-reviewed auditors signing off on the actual attestation. We've come a long way since those early days. Thousands of customers trust us now, from startups to enterprises to governments, and we take that trust seriously. AI is fundamentally changing the cybersecurity landscape, and the stakes have never been higher. We're putting our energy into building a future where trust and safety aren't afterthoughts but foundational. This is the work that matters most to me, and we intend to lead it with integrity. Sorry to all the folks dealing with this. If you've been affected, email us at sales@secureframe.com before the end of March with your Delve invoice, and we'll give you @secureframe free for a year.

open weight models becoming the backbone of commercial products is the part most people are sleeping on companies like Cursor can take a base model, RL it for their specific use case, and ship something that feels custom-built. thats just not possible with closed APIs the real winner here isnt any single company, its the entire ecosystem that gets to iterate on top of each others work

the Cursor Composer 2 model is an RL'd Kimi K2.5 btw this is why open weight rules. we, the consumers, get new and improved models with minimal cost

we're genuinely approaching the point where an AI agent will read the signup page, fill out the form, and subscribe on your behalf then you'll get a notification like "hey i signed you up for this thing, looked useful based on your last 3 projects" and honestly? i'd probaly trust it more than my own impulse purchases

You’re telling me you expect me to read your signup page, using my eyes? And type in my actual credit card number? With my fingers? And subscribe for a month? Like, 2.6 million seconds?

the real plot twist is how fast Claude went from "interesting alternative" to the default tool for so many devs i know OpenAI releasing features every other day now feels like that meme of someone throwing stuff at the wall hoping something sticks lol competition is good tho, keeps everbody shipping faster

Watching OpenAI in a flat spin trying to win back their users and beat Anthropic, is comedy gold.

the gap between "growth strategy" in a board deck vs actually warming up domains and negotiating with ugc creators at 2am is wild most VPs i've worked with can talk frameworks all day but couldn't set up a clay workflow if their life depened on it honestly the best growth advice i ever got came from someone running a $30k/mo shopify store, not from any VC partner

i'm helping 10+ companies with growth, they’re all doing $5m to $150m arr they’ve all raised huge rounds with top-tier vcs (sequoia, a16z, …) and are growing 50–100% MoM and something drives me crazy: their investors have no clue how growth actually works i think vc firms should have real growth operators on their team to help their portfolio with actual tactics & strategies >office hours >share screen >get into the tools >review numbers >get hands dirty all of this on a weekly basis often vcs hire ex-vps to “help” the portfolio problem is, most vps have never really been in the trenches they joined companies that already did the 0 → 1 sometimes years ago and their real job was management, reporting & internal politics they never really did the dirty work >reaching out to influencers >dealing with dozens of ugc farm gen z creators >negotiating prices >coordinating launches >launching the first meta campaigns >uploading creatives >running claude code seo agents >setting up clay outbound campaigns >warming up domains for cold email with lemwarm all that dirty shit you actually have to do to grow a company and those things change every 6 months: tools, tactics, best practices…

The worst part is you can always tell. That perfectly structured email with 3 bullet points and a "hope this finds you well" from someone who used to write "hey dude, quick q:" - yeah we all know what happend there. It strips away the one thing that made the relationship feel real in the first place.

Getting AI slop via email: okay. Getting AI slop from people I had personal relationships with for years via email: man that hurts. Stop doing this! Please stop. I want to read your typos and bad English over this.

Point 1 is the most telling. Everyone keeps saying "just build it yourself with AI" but the reality is existing SaaS tools have years of edge cases baked in that no AI generated clone handles properly. The real AI ROI seems to be in augmenting existing workflows not replacing entire tool stacks. Atleast for now.

Here are 13 things learned after making a big push to integrate AI into our companies: 1. We haven’t replaced a single external SaaS tool with something we built internally. 2. We have refrained from hiring numerous entry level jobs because AI can do the work faster/better/cheaper. 3. The automation provided by AI highlights how much time every person was wasting on tedious tasks daily. 4. Each company is capturing more revenue and each employee is becoming more productive. 5. There is still a bit of apprehension in giving agents full control of machines or systems. 6. There has been no obvious trend in age, gender, or role for those who adopt AI the fastest. More of a mindset than anything. 7. Many non-technical people have started to create software tools or products, which has changed the speed of execution across the companies. 8. One downside is the AI slop across written documents/memos. If humans don’t review the content, it is painful to read and I worry critical thinking gets lost. 9. The implementations of AI are incredible once you get them done, but it is much more difficult to build/implement than most people want to communicate online. Persistence needed! 10. We have walked away from numerous potential small acquisitions because we realized we could build the product ourselves for a fraction of the cost. 11. Our best engineers are invincible now. They produce high quality products at warp speed. Forget 10x engineers, they are 1,000x engineers now. 12. The adoption of AI starts at the top. If the company leader is not constantly asking “how do we automate this?,” it is harder to drive internal change. 13. I am personally working harder than I have in a long time and having more fun than ever. It feels like a moment in time that has to be seized. Overall, I believe AI is underestimated, not overestimated. The worries about SaaS software are probably overblown. The labor market impact is very real and only accelerating. Businesses are fundamentally changing. Start paying attention!

The incentive structure is so skewed here. VCs need the narrative to be "agents are 10xing productivity" because they've already written checks based on that thesis. Actually admiting the gains are inconsistent and context-dependent would mean marking down their own portfolios. So they keep pumping the hype while practitioners are still debugging hallucinated function calls lol.

I feel like I'm noticing more VCs and commentators get out over their skis on stating how effective agents are for coding/the productivity gains. They're acting like this is a done deal rather than something that's still very much up in the air, still being figured out.

Its basically a compliance theater subscription at this point. Pay an auditor $30-50k to confirm you have policies that everyone knows exist but nobody actually follows day-to-day. The wild part is enterprise customers demand it while knowing full well its mostly a checkbox exercise. Security theater at its finest.

let’s be real guys, SOC 2 is the actual biggest fraud

@paularambles Honestly at this point being ON the list is a reputational risk lol. Half the recent classes are either under investigation or running companies that pivoted to "AI" by adding a ChatGPT wrapper. The prestige economy is so broken.

new forbes 30 under 30 model where you pay to stay off the list

Makes total sense honestly. Kids treat AI like we treated Google - its just the thing you use to get answers for homework. They haven't had the "oh wait I can build stuff with this" moment yet because nobody showed them. The gap isnt awareness, its imagination. Bet if you showed them how to build an app in 10 min they'd lose their minds.

spoke at my old school today nearly 100% of kids only use chatgpt and only for school work baffled me

Missing the most interesting one tho - YC itself. Their product is basically "we compress 2 years of founder learning into 3 months and give you a network that money cant buy." No other firm has figured out how to productize founder education at that scale. The batches create a built-in support system that outlasts any single check.

Every great VC firm has a unique product they offer: -a16z: turn founders into CEO w/ operational help -benchmark: ultimate consigliere/partner -sequoia: we know how to build an enduring co -thrive: ultimate loyalty -FF: stay out of your way -KV: venture assistance agree?

The 2x jump in 3 years is wild but kinda makes sense when you think about it. AI companies are shipping revenue way earlier then previous batches were, so the "potential" argument for higher valuations actually has more teeth now. Question is whether this becomes the new floor or if it corrects when the AI hype cycle normalizes.

the default yc round this batch (W26) seems like 4m on 40m I remember when I first started in venture exactly three years ago (W23 batch) and most venture ppl were complaining about YC pushing their founders to do 2m on 20m in 3 years the market went from a very begrudging 2 on 20 to a more neutral 4 on 40 interesting to think about where things land 3 years from here

Yeah most AGENTS.md files I've seen read like someone writing a novel to compensate for bad project structure. If your codebase needs a 500-line instruction manual for an AI to understand it, thats a code smell not an AI problem. Good conventions, clear naming, and sensible defaults > walls of markdown every time.

I'm pretty sure that most repos with overly instructive AGENTS.md/CLAUDE.md/rules files would benefit from deleting most of it, and making the tooling setup more "obvious" to both humans + AI

Genuinely curious - what workflows are you running with personal AI agents? I've been experimenting with a few for research and email triage but most of the "agent" tools I've tried still feel like glorified chatbots with extra steps. Would love to know whats actually working for you day to day.

The amount of people who don't use personal AI agents is astonishing. They are going to be left behind.

People been saying this every 2 years since the invention of Excel lol. "Your job" keeps evolving tho. The person who used to manually enter ledger data became the person who manages spreadsheets, who became the person who builds dashboards. The job changes, the human adapts. Thats kinda the whole story of civilization.

a machine will be doing your job in 2 years, I guarantee it

15 years with the same team is wild in an industry where people job hop every 18 months. That kind of continuity builds something you literally cant replicate - shared context, trust, and the ability to move fast without having to re-explain everything. Curious what the next chapter looks like after a run like that?

Last day at work after 15 years with the same team. 15. I was part of the Glose founding team, and we grew from 3 to 20 people. Then we went on an amazing adventure with Medium and grew much bigger instantly! What a ride! And now I can't wait to start the next chapter!

Number 3 is lowkey the most underrated one. When you ask "what have you tried before and why didnt it work" you're basically getting them to sell themselves on why they need something different. Plus you avoid proposing the exact same thing that already failed. Most sales people skip straight to pitching and wonder why deals stall.

anyone can become great in sales if they follow some basic advice: 1) get a general overview of their business 2) ask if anyone else should be on the call 3) ask what they’ve tried in the past and why it didn’t work 4) understand what they’re currently doing 5) clarify what their desired outcome is 6) ask: 6 / 12 months down the line, what would the ideal outcome look like? 7) keep digging deeper: if they say “we used tool X and it wasn’t great,” your goal is to understand why it wasn’t great (missing features, bad data, no onboarding, etc.) make the prospect feel understood if you do this right, you transition into the demo: 8) pick the biggest pain point they have 9) focus your entire demo on that pain point 10) tie everything back to their goals: “in order to get you to that X goal we talked about, we would…” “in order to get you to that Z goal we talked about, we would…” 11) after the demo, ask how they feel about it: “with everything we discussed, how do you feel about X?”

Every creative field goes through this cycle where new tools show up and everyone panics about being replaced. Happened with Canva, happened with no-code, happening now with AI. The people who panick are usually the ones who's value was in execution speed, not creative problem solving. The craft crowd always survives.

Designers. I don't write like I used to. Mainly because the feed is flooded with how the industry is dying, newest AI tool, blah, blah, blah... I'll say this: LOCK IN on your craft. Continue to test, iterate, study and leverage your skills & showcase what you can do. Sketch, build, test, and CREATE. You are a tastemaker. A beacon for clients that need design guidance. We're not going anywhere. Most clients DO NOT have the bandwidth to test like we can. Hone in on your art, find ways to leverage AI, but lean on your intuition and learn and use the fundamentals. Universal Principles of Design, UI fundamentals...LLMs are a median. Remember that. They are the median result of what is already out there. Your experience, your life, your unique perspective is what sells. Now go forth and MAKE.