Anton Osipov

I’ve decided to stop work on Bobr AI (AI for slides) It was my first big product after the success of my $3M ARR company and it also was my first global product. I spent $20k on marketing to make $8k, and I realized a few things: 1. You can’t win in a crowded market without a huge budget or a non-trivial go-to-market strategy. I couldn’t crack distribution. 2. Product quality alone is not a game changer. 3. Investors don’t believe in you if they don’t believe you can become a winner in the market. You can spend your own money (like I did) and grow to tens of thousands of dollars in MRR, but that’s not the path to building a unicorn. So now I want to build a unicorn keep testing hypothesis and share the journey here

26k on threads vs 85 on X. Alright, I’m packing my bags.

Omg, I just started posting on Threads 2 days ago and already got 2.3k views. Threads looks better than X. What do you think?

Day 6 after I closed my startup. Today is Saturday. Yesterday I still shipped a small feature in Bobr AI even though the product is already closed, because one user asked me for it and I couldn’t just ignore him. Today is my first weekend without a product to build. I’m not going to work on anything today, but it’s going to be hard.

@nestymee Thank you for sharing

⚡️ X algo update, tldr for creators: - Optimize for the bundle: replies, reposts, profile clicks, not just likes - Negative signals (mute / block / report / not interested) carry negative weight. Engagement-bait that backfires is worse than no post - Author diversity penalty is real - 10 posts in 10 min and the last ones get attenuated - Hashtags are dead. Out-of-network reach is now learned via embedding similarity - Profile clicks + follows are first-class predicted actions. Strong hook + bio that converts compounds - Thunder guarantees your followers see recent posts - recent window only. Cadence matters

Apparently Bing saw my post on X about shutting down bobr ai and instantly sent it to the graveyard of search results.

#day5 Feeling when you shut down your startup but still get subscription payment notifications

@TimHaldorsson Thank you

yesterday we had full house at our Codex community afterwork 🇵🇹 it was little over 100 builders thats came to our office to listen to 3 community demos and a panel focused on why people are going for Codex. it was good vibes and many good conversations during the evening. we are planning our next Codex event, so if you have a cool demo or have built something great with Codex, reach out and hopefully we can highlight it for the next event. thanks to @OpenAIDevs and @reach_vb for making Codex a great AI building platform.

Day 4 since I shut down my startup for AI slides. Today I had a great conversation with the CTO of @truewindai. I applied yesterday for one of their roles, but during the conversation he said I seemed more like a founder than a typical hire and encouraged me to build in the US The team is building a really impressive AI system for accounting automation. I liked their approach a lot especially the focus on deterministic workflows and integrating with how accountants already work instead of forcing completely new behavior. I also visited a Codex event by @EspressioAI . It was interesting, but I’m still stuck with Cursor 😂 Didn’t work on new ideas today, but I found a few hypotheses worth testing. Keep in touch.

Codex Community Afterwork

@VeronicaBuron I’ll send you in DM

@MeOsipov can i have it

Got $50 of Cursor credits from a Cursor event, but I can’t use them since I’m on my company’s subscription. Giving them away to random people in the comments.

@elgermerlo @X @Reddit @LinkedIn Google ads

What's your choice on distribution? @x vs @reddit vs @linkedin

Day 3 since I closed my AI-for-slides startup. Updated my resume. Not because I want a job. Just in case the idea I'm waiting for never shows up. Writing it, I realized I couldn't remember half of what I'd built over the years. So I ssh'd into a server that's been running for 5 years and counted nginx vhosts. 75. Most of them I'd already forgotten existed. I'm not afraid of starting. I'm afraid of starting wrong again. Because after 75, you're not looking for "an idea." You're looking for the one that's worth not having a 76th. Btw, write your resume even if you don't need a job. It shows you that none of it was wasted. Even when you have no idea where to go next.

@nico_jeannen Same happened to me with a designer. Now I hire only through recommendations

Im never hiring ppl from X again 💀 Tried hiring motion designer for a video for a 30s video > Found someone on X who shared nice videos > Slightly more $$ as other quotes I received, but portfolio looked good > I want to be easy to work with, so I pay in advance, no hard deadline, trusted him to do its job properly Result: >0 communication, have to babysit every little thing >garbage result I can't use > ok maybe instructions not clear enough? > Sent exact mockup of what I needed > Still didn't follow what I asked for 💀 Lesson learned lol, I'll use Fiverr/Upwork next time The bar is so low, why can't people just do what they paid for? How do you find people who can actually do their job and follow instructions? 🥲

@CiprianiRanieri It’s impossible to scale to big numbers without clarity. I already have a company making tens of thousands in revenue, and I wanted to build something big that could win the market. But I didn’t find PMF, so I abandoned it.

@MeOsipov why?

Founders only. Drop the side project you abandoned. I'll tell you if it's worth resurrecting. ↓

@tibo_maker People don't like AI Slop

looks like a fail ❌ those 2 video posts featuring Pip didn't do that great please tell me, should I: 1. give up on the mascot idea 3. change the mascot 2. do more, it's fun 4. actually ... 👀

@krishmodi404 @PalantirTech hb 🥳

turned 21 today big year ahead: > joining @PalantirTech in NYC > built the fastest scheduler for LLM requests > launching something in the next couple days to make all agent systems run faster and cheaper couldn’t be more excited for what’s next

Day 2 without a project. Ten years without a day like this. I thought it would feel like rest. Feels more like phantom limb — body still reaching for notifications, dashboards, the next thing to ship. Spent today pitching ideas to Claude. It agreed they were great. I told it they were nonsense — it agreed with that too. The work wasn't only the burden. It was also the scaffold.