Jamal Aghayev (@jamalaghayevin) header
Jamal Aghayev (@jamalaghayevin) profile picture

🏴‍☠️ co-founder and ceo - @cuprice_io Venture Partner - Tenity UK My portfolio: 5 startups, 2 exits, 700+ supported startups, 60+ acceleration programs.

Joined May 2020
Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)Photo by Jamal Aghayev (@jamalaghayevin)
Half agree. Big AI can copy any feature. It can't copy your customer. @levelsio says niching down doesn't protect SaaS anymore, because the models can now do everything. True for niches built on a feature. A tool that does one task is a prompt now. Not true for niches built on knowing one type of customer better than anyone. Their workflow, their weird edge cases, who they call when it breaks. In my view, niche down on the customer, not on the feature. Which one is your niche built on?
💯 There is no real way to survive for most SaaS companies here because AI is so completely different than any technological advancement before it Before you could just niche down and specialize on one thing and you'd know for certain the big tech companies would never be able to dedicate enough resources to fill your tiny little niche market of $300M where you could take 1% of the market and make $3M/year Riches in the niches! And so you'd avoid going too general, because "you'd spread yourself too thin" and hit "death by a thousand paper cuts" And general was where big tech companies were great at, they could dedicate huge resources in solving these big problems, like building Google Sheets, or Salesforce (whatever it is they do) But that was the old days, and it's completely different now With AI, big tech companies can easily fill any niche, or be general at the same time, they can do the whole range actually, without even spending time on consciously doing that, because they're not offering a particular SaaS offering at all They're offering you artificial general intelligence (AGI) that can do it all And as long as they improve their models every time and make them bigger and better, it slowly can do literally anything (okay not yet anything, for anything they'll need physical robots, but they're working on that too of course) This leaves SaaS companies with a bitter ending: - they either do nothing and stay in their niche like they economically should, but that doesn't work anymore now - they try go general, and offer EVERYTHING and ANYTHING, and that works for a little bit until customers realize they might as well use whoever makes the models that the SaaS itself uses anyway, which is the big tech and big AI companies, who by the way either offer it free or for much cheaper Everyone's been checkmated by big tech and big AI This will be the creative destruction of the century But there'll be some survivors and then a lot of new companies, but how they'll look nobody knows yet!
1
1
25
Jamal Aghayev retweeted
The First 1000 Customers Playbook Very excited to announce our second multi-company playbook, that takes 6 top-performing companies (@Lovable, @canva, @WisprFlow, @higgsfield , @mercury , @gammaAI and deep dives into how they acquired their first 1000 customers. Attached is the first rule from the playbook. Check out the link below is to the entire playbook. Founders: hope you find it helpful as you think about finding PMF.
4
22
5
254
61,123
Dear early-stage founders, are you hiring because you need help, or because you're tired? Tired is not a job description. Before a first hire, I'd want 4 answers on paper: 1. What will this person own, start to finish? Not "help with marketing". 2. What have you done by hand for at least a month that they'll take over? 3. How will you know in 30 days that it's working? 4. Can an AI agent do most of it today? If yes, try that first. It's cheaper to fail. My take: your first hire should take a job you already know how to do. Not a job you hope someone else will figure out. So, help or tired?
1
44
Jamal Aghayev retweeted
2026 Startup Bookshelf 📕
5
10
1
163
9,736
Before you put a price on any AI feature, answer 3 questions. What does one run cost you on a bad day, not an average day? What does one good outcome save your customer? Can your customer predict the bill before it arrives? Most founders only answer the first one. Then they wonder why customers churn after the first invoice. I call the third one the NO SURPRISE RULE. A fair price that scares people is still a bad price.
1
1
38
12 frontier models tested. The winner is also 17x cheaper. Sol did the job at $0.11 per task. Fable 5.2 did it worse at $1.87. Fable 5.1 costs 430x more than GPT-6 Luna for similar results. But the line I care about is the last one. When agents help where nobody asked, that extra work costs about 80% more time and money. That is the pricing problem nobody models. Your cost per task is not fixed. It depends on which model you pick this quarter and how often your agent does work nobody needed. If your price assumes today's model costs, it expires with the next release.
What’s the best model for proactive agents? Today, we’re introducing the Agent Initiative Benchmark. We tested 12 frontier models on realistic workplace tasks inside a synthetic company with 6 years of highly detailed history: it has a financial model, CRM records, past-due invoices, Linear tickets, and real-life events affecting company development - competitor product announcements, SVB collapse, etc. We used the company’s Slack to test how a proactive agent listening in on each conversation would assist the team in day-to-day tasks: from stepping in to align initiatives across teams to warning the team not to use an unapproved customer logo. Highlights: - 6.1 Sol and Astra from @OpenAIDevs beat Anthropic's latest Fable 5.2, with Sol crushing it at both performance and cost ($0.11 per task vs $1.87). - The most expensive models aren't the best: Fable 5.1 is 430x more expensive than GPT-6 Luna, with similar performance. - Most models try to help even if they're not needed. Interestingly, in these failure cases, they spend more time thinking, and the unnecessary work takes about 80% more time and money than if they stayed quiet. If you’re building agents, keep reading for more tips on which model is best for your use case. Detailed breakdown, the open-source dataset, and benchmark code below:
1
1
54
Building @cuprice_io , I keep seeing the same thing. The founder knows their churn, their CAC, their burn to the dollar. Ask when they last changed their price and they go quiet. Your price is the only growth lever you can ship in one afternoon. And it is the one most founders never touch. Stop treating your pricing page like a tattoo.
3
1
2
49
Seat based pricing is a tax on your customer's success. You sell an agent that does the work of two reps. The customer cuts two seats. Your revenue drops because your product worked. That is the SEAT TAX. And this week the 20VC x SaaStr crowd said out loud that seat based software can't get a bid. If your product replaces work, charge for the work. Not for the chairs.
22
Jamal Aghayev retweeted
a few YC companies have been wrapping our API and doing millions in revenue off this feature now it’s available for everyone
Today, we are launching person verification on Crustdata Know if a profile is real before anyone books an interview Crustdata has grown to $10M ARR by powering some of the fastest growing companies and enterprises, including Wispr Flow, HeyGen and many more Now, it's time to stop trusting every headline We searched for Senior Software Engineers at Google and found 10,202 profiles. 32 of them have "Student" in their headline. Some are real engineers studying on the side. Some might not work at Google at all With person verification, send a profile and we check: > Do the dates add up > Do real coworkers engage with their posts > Is there independent evidence of the role And you get a verdict, the reasoning and the sources
2
4
66
4,757
Jamal Aghayev retweeted
36 questions every VC should ask to understand if a founder has what it takes to succeed - how fast do they execute? - do they have good judgment? - can they handle suffering? - are they highly intelligent? - are they comfortable with ambiguity? - are they driven, headstrong, and bold? - do they think independently and challenge conventional ideas? - are they disagreeable and comfortable not pleasing everyone? - are they neurodivergent? - did they experience childhood trauma? - are they on the autism spectrum? - are they 100% committed? - can they handle a psychologically intense environment? - are they obsessive and able to focus? - do they have grit, determination, persistence, and resilience? - do they fold, freeze, or lie when things get tough? - do they think and speak in numbers? - do they know their most important metrics? - do they have enough technical knowledge? - are they good storytellers and salespeople? - are they charismatic? - are they good at fundraising? - can they recruit and hire people better than themselves? - are they good at firing? - are they intensely competitive and driven to win? - do they have something to prove? - do they learn fast and adapt? - were they entrepreneurial as a child? - what did their parents do? - do bigger, crazier visions excite them? - are they well-connected or outside established networks? - do they come from a center of excellence? - have they started a company before? - can they answer questions concisely and clearly? - have they thought through every aspect of the business? - are they confident enough to change their minds and accept feedback?
You’d think we’d have learned from Steve Jobs and Travis Kalanick that one does not just simply replace the founder…
9
13
2
58
4,552
Jamal Aghayev retweeted
Replying to @inmerge_az
@inmerge_az returns to Baku on 30 November–1 December 2026. 🚀 🇦🇿 The summit will bring together more than 200 international speakers from technology, business, entrepreneurship and innovation. For startups, INMerge 2026 offers the chance to represent at the Startup World Cup 2027, with 40+ startups competing in INBattle for a €45,000 cash prize. Now in its sixth year, INMerge continues to connect startups, investors, corporations and ecosystem leaders across Europe, the Caucasus, Central Asia and the Middle East. eu-startups.com/2026/11/inme…
6
4
357
He is right!
Revolut's CEO says a product team can now be two people. He still runs teams of 10. Nik Storonsky, CEO of Revolut, on AI and team size: 1. You no longer need separate iOS and Android developers 2. One developer can cover frontend, backend and maybe data science 3. With today's tools, even some of the design 4. Revolut is organized by specialization, with separate systems for each role 5. So it keeps the old model until the new one is proven The plan: automate first, then hire fewer people. A startup has nothing to undo, so it can start with teams of two today. Full breakdown here: productmarketfit.tech/p/revo…
30
Omg so dark 🥲
claude woke up and decided to have granola for breakfast today
30
Jamal Aghayev retweeted
" You're gonna be remembered forever because some things are eternal."
430
2,636
473
23,407
7,799,239
Excellent!
In 2014, I sold my first company Loom to Dropbox. Today, I’m proud to announce Roma 2.0 (YC F26): the first to-do list for AI agents. Everyone uses a different agent: Claude, Grok Bot, Muse… Until now, none of them could work together. Roma automatically creates tasks and hands them to the right agent. Allowing you and your agents to work together, with shared context. Try it here: roma.app
34
I don’t do my to-do list anymore. I send a voice note. An AI agent does the work. I only review the result. I’ll show how I built it, what broke first, and the simple rules that fixed it. 3 minutes, one real system. TechBrains Baku 2026 is one of the biggest tech community events in Baku. 100 speakers, 10 tracks, 2 days. "How to Become an AI-Native Founder" 21 and 22 November, Baku.
1
4
87
A couple of years ago I was one of the people who first brought @sgbakuhub, as chapter co-director. Now it's back. On Nov 16 I join the new team's first event, this time as the fireside chat guest. Loc: Holberton School, Baku - 7 PM startupgrind.com/events/deta…
1
40