NeverKnower.base.eth @neverknower_dev

web3 developer, launching @say_more_gm and TactiBall on Base you'll never know until you try 👀 Joined May 2024

Tweets

852
Followers

163
Following

413
Likes

2K

NeverKnower.base.eth @neverknower_dev

2 weeks ago

@altagers yeah, it sounded cool at the time, and looked like the beginning of a cool story.. but it turned out to be the end of it 😢

0 0 0 12 0

View Details

NeverKnower.base.eth @neverknower_dev

2 weeks ago

@iamdevroyale Let’s build! 💪

0 0 1 7 0

View Details

NeverKnower.base.eth @neverknower_dev

2 weeks ago

GM! it's a bit weird, but I like to see Bitcoin is down it means it's still good time to build, and I'm not too late. Bitcoin 100% will go to new ATH, it's just matter of time. And I hope I have like 3-6 month until that. So I'll be with ready product at that time. Let's keep building!

2 0 2 43 0

View Details

NeverKnower.base.eth @neverknower_dev

2 weeks ago

@vbm_177 And then every new -5k.. 😅 DCA always win long term!

0 0 1 34 0

View Details

NeverKnower.base.eth @neverknower_dev

2 weeks ago

gm! so, will BTC break 60k and go below? what do you think? 👀

1 0 2 46 0

View Details

NeverKnower.base.eth @neverknower_dev

a month ago

looking at how people are canceling subscriptions on Claude Code, and disappointed that spent too much time on Claude Code stack. Well, vendor lock is never a good idea. For now, Codex is the best choice, but they can adjust limits any time after they beat Claude Code.

0 0 1 47 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

@_clemens__ @base DMed you. Would be happy to join 👌

0 0 1 16 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

building AI company rn. Have the similar vision - I don't want to do "autonomous AI company", I want agents to take care of 90% of work, and I'll be able to do that 10% much better, I'll have time for marketing finally and a lot of really important things apart of just building. The hardest part for me during last few weeks - setting up all this takes time. A lot of pitfalls, errors to solve, time to waste on something that doesn't work as it should. It's not like 1-click and it magically works. But it 100% worth it and I'm finishing this part 💪

0 0 3 23 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

@altagers check Alibaba Cloud Models. That's the best for now by price/value.

0 0 1 15 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

GM! I'm still here and building. Last few weeks I'm working on setting up AI agents. My goal - speed up project developments, delegate most of routines to agents. It's not magic button that will do everything for me, but 80-90% of work - quite possible. Will share my insights and progress here 👀

0 0 1 31 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

oh, a lot of tbh. - @say_more_gm - one new project to monitor token launches through ContinuousClearingAuction introduced by Uniswap - TactiBall - planning to get back to it once I'll have more time - and one non-crypto project But for now I'm playing with AI, to setup agents to help me build all it 👀

0 0 3 26 0

View Details

NeverKnower.base.eth @neverknower_dev

4 months ago

gm! Long time no see. How is it going? I see a lot of posts about AI, but quite few about crypto. Is the crypto still alive? I'm still here, building next cool project that will help with new token launches! Also, actively working with AI, setting up agentic coding pipeline - works nice 🤖

2 0 4 41 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

@austingriffith @AnthropicAI @escottalexander @_carletex_ that's sweet! thanks for the work and for sharing it :)

0 0 1 72 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

@antigravity well, started to receiving this today. Is that "sharing capacity" already or not? 😅

3 0 20 2K 0

View Details

NeverKnower.base.eth @neverknower_dev

3 months ago

@0xDesigner well, disagree. Vibecoding originally was about using AI by non-technical people. Building agentic coding pipelines requires skills, that non-tech people lack. So, that's different!

0 0 0 49 0

View Details

NeverKnower.base.eth @neverknower_dev

4 months ago

@CasperOnChain GM GM! How is it going here in CT? 👀

0 0 0 4 0

View Details

NeverKnower.base.eth @neverknower_dev

4 months ago

@GrafinkaVi Which city? 👀

0 0 1 68 0

View Details

NeverKnower.base.eth @neverknower_dev

4 months ago

@KamilStanuch Damn, missed this! When the next one? 👀

0 0 1 24 0

View Details

NeverKnower.base.eth @neverknower_dev

4 months ago

Disagree with that. I'm using LLMs every day(Codex 5.3 now mostly), and they help a lot and save a lot of time. But that's not AI. It cannot think and make good design decisions. If we're talking about something more complex than new habit tracker. And I'm expecting this AI hype will get stuck quite soon, cauze current approach has limits, and it's not easy to bypass them quickly. So yes, new models will come, they will be like 5-10% "smarter" than previous, but to do next big leap - we need to use different approach than LLMs. Most probably. And it takes years to achieve that. So, when people will understand that - AI market will crush, cauze expectations are so high rn. That's how I see it. It doesn't mean that current AI is useless. I agree on most points in the article - you should already use it and integrate in your life. It just boost your work by 5-10 times (really), and if you're not using it - you're losing to your competitors. That's correct point. But I don't believe we're near something really big in next few years, mostly opposite - we're near to face the reality, which is - it's not AI yet.