Azamat K.

Introducing @siriusai. AI retention model. We’ve proven that trust is more profitable than friction. Since August, we’ve helped companies like Promova generate millions in revenue. Comment Sirius to get an exclusive look at the platform ✨

Early-stage founders love optionality. "We can go B2C or B2B." "We can go upmarket or down." "We serve three verticals." Optionality at the early stage is a tax on speed. Pick one lane. Win there. Expand later. We did this with Sirius. Consumer subscription brands. Retention. AI. That narrowness is the reason we move fast.

I spent two years in college. The only useful skill I picked up was networking. So I dropped out and moved to San Francisco to build. Two years of classes. Zero applied to my startup. Every connection I made in SF applied immediately. The most expensive lesson is figuring out where you learn fastest.

Sirius retains 20%+ of churning subscribers. Zero human intervention. The system reads subscriber data, identifies why someone is leaving, and deploys a personalized save in real time. This is what AI should look like. A measurable outcome you can point to in a board meeting.

Adding a "pause subscription" option to your cancellation flow recovers 10-15% of churning subscribers on its own. A huge chunk of cancellations come from temporary situations. Travel. Budget crunch. Seasonal changes. Give them a bridge back and they'll walk across it.

Every dollar of acquisition is a tax on weak retention. And the bill always comes due. Subscription companies LOVE to talk about CAC like it is a fixed cost of doing business. Spoiler: It almost never is. CAC is mostly a function of how leaky your bucket already is: ↳ Keep 95% of your customers… …and you can spend less to acquire because the ones you have are still paying you next year. ↳ Keep 60% of your customers… …and you have to spend more every quarter to stand still. It’s like the business is a treadmill. The growth team has to run on it. But the retention number is what gets to set the speed. Founders feel this long before the data makes it obvious. The signs show up in a specific order every time: > First: payback periods stretch by a month, then two > Second: a board meeting where someone asks why ad spend keeps climbing while ARR looks flat > Third: layoffs Acquisition is the symptom that gets the budget. Retention is the cause that gets the leftovers. Reverse that order and the math behind every other line item starts working in your favor.

Every market hypothesis is a guess until someone hands you money. Each time, the market I thought I was building for shifted the moment the first customer showed up. Customers teach you your market. Spreadsheets never will.

Focus. Focus. And again Focus.

"We're sorry to see you go" saves almost nobody. A personalized response based on usage, billing history, and cancellation reason keeps 4x more subscribers. The data already exists in your stack. You're sitting on recovered revenue and ignoring it.

ideal friday event: a hands-on technical workshop or casual coffee meetup?

who's an under-the-radar person in the bay area doing incredible work right now?

be honest: do you actually learn from panels or do you just go for the mixer?

what subculture or industry community is sf totally missing right now?

think about the coolest or most useful professional connection you made in sf this year. how did you two actually meet? structured mixer, casual coffee, accidental run-in...

what’s a random, highly specific niche topic you’ve been deeply hyper-focusing on lately? like something you could easily talk about for 2 hours straight if someone asked

what was the last meetup, panel, or event you went to in SF where you actually stayed until the very end?