aditya's banner

aditya

@adxtyahq • 29,529 subscribers

21 | DevRel @EntelligenceAI | DM for Work/Collab

Shorts

bro. someone built a terminal weather app in Rust with live ASCII animations. it literally rains / snows inside your terminal when it’s raining / snowing outside 😭 this is peak software.

bro. someone built a terminal weather app in Rust with live ASCII animations. it literally rains / snows inside your terminal when it’s raining / snowing outside 😭 this is peak software.

807,781 görüntüleme

“16 million exchanges?!” “This is called open source”

“16 million exchanges?!” “This is called open source”

697,632 görüntüleme

GOOGLE HAS BEEN FAILING PRETTY BADLY TO PUT OUT A MODEL THAT CAN ACTUALLY HANG WITH THE BEST LATELY tested Gemini 3.8 Flash against Kimi K3 on the same Sticky Ball game Flash was insanely fast and used way fewer tokens, but the actual game was nowhere close. K3 had much better mechanics, movement, and overall game design Flash clearly has the speed and efficiency part down, but the gap in what it can actually build is pretty big

GOOGLE HAS BEEN FAILING PRETTY BADLY TO PUT OUT A MODEL THAT CAN ACTUALLY HANG WITH THE BEST LATELY tested Gemini 3.8 Flash against Kimi K3 on the same Sticky Ball game Flash was insanely fast and used way fewer tokens, but the actual game was nowhere close. K3 had much better mechanics, movement, and overall game design Flash clearly has the speed and efficiency part down, but the gap in what it can actually build is pretty big

135,757 görüntüleme

just asked Grok 4.5 Fast inside Cursor to "make a weather app", taking inspiration from the video they posted it's totally shit and doesn't match the demo in any way that demo is incredibly misleading.

just asked Grok 4.5 Fast inside Cursor to "make a weather app", taking inspiration from the video they posted it's totally shit and doesn't match the demo in any way that demo is incredibly misleading.

177,159 görüntüleme

I'M PROBABLY ONE OF THE PEOPLE CONTRIBUTING TO THE GPUS DYING 😭 Built the same Crossy Road clone with both Kimi K3 and Opus 4.8 using their respective API keys inside Cursor, so neither model had a framework advantage despite the 2.7× difference in cost. Both produced fully playable games with solid physics, but Kimi's UI felt slightly more polished Kimi cost $7.11, Opus cost $19. Pretty impressive considering the price gap

I'M PROBABLY ONE OF THE PEOPLE CONTRIBUTING TO THE GPUS DYING 😭 Built the same Crossy Road clone with both Kimi K3 and Opus 4.8 using their respective API keys inside Cursor, so neither model had a framework advantage despite the 2.7× difference in cost. Both produced fully playable games with solid physics, but Kimi's UI felt slightly more polished Kimi cost $7.11, Opus cost $19. Pretty impressive considering the price gap

47,488 görüntüleme

small dev experiment today came across Cal AI (recently acquired by MyFitnessPal) 15M downloads, ~$30M/year tried vibecoding it myself on superapp wasn’t gonna pay for the subscription took a few prompts and I had a working v1 lowkey crazy how fast this works

small dev experiment today came across Cal AI (recently acquired by MyFitnessPal) 15M downloads, ~$30M/year tried vibecoding it myself on superapp wasn’t gonna pay for the subscription took a few prompts and I had a working v1 lowkey crazy how fast this works

35,980 görüntüleme

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

I WAS WRONG, FABLE-5.1 MIGHT ACTUALLY BE THE BEST MODEL TO EXIST ON THIS PLANET FOR FRONTEND TOO 😭 Tried recreating the exact same game with Fable-5.1 and Kimi K3. Fable-5.1 outclassed Kimi K3 on UI and gameplay. Smoother mechanics, better interactions, and noticeably more polished overall. Kimi was still competitive, especially with way fewer jitters during gameplay. Cost: - Kimi K3: ~$11 - Fable-5.1: ~$18 That’s ~60% more cost for Fable, but the quality difference was actually noticeable. Kimi is still ridiculously competitive for the price.

I WAS WRONG, FABLE-5.1 MIGHT ACTUALLY BE THE BEST MODEL TO EXIST ON THIS PLANET FOR FRONTEND TOO 😭 Tried recreating the exact same game with Fable-5.1 and Kimi K3. Fable-5.1 outclassed Kimi K3 on UI and gameplay. Smoother mechanics, better interactions, and noticeably more polished overall. Kimi was still competitive, especially with way fewer jitters during gameplay. Cost: - Kimi K3: ~$11 - Fable-5.1: ~$18 That’s ~60% more cost for Fable, but the quality difference was actually noticeable. Kimi is still ridiculously competitive for the price.

367,938 görüntüleme • 1 ay önce

someone ran a speed test inside NVIDIA GeForce Now and got ~5000 Mbps what am i even paying for-

someone ran a speed test inside NVIDIA GeForce Now and got ~5000 Mbps what am i even paying for-

1,426,306 görüntüleme • 5 ay önce

OFFICIALLY SAYING KIMI K3 IS BETTER THAN CLAUDE FABLE-5 AND GPT-5.6-SOL FOR FRONTEND DEVELOPMENT Tried building a macOS-style UI for an AI operating system with Kimi K3. This whole thing took about 15 minutes from a single prompt.

OFFICIALLY SAYING KIMI K3 IS BETTER THAN CLAUDE FABLE-5 AND GPT-5.6-SOL FOR FRONTEND DEVELOPMENT Tried building a macOS-style UI for an AI operating system with Kimi K3. This whole thing took about 15 minutes from a single prompt.

556,795 görüntüleme • 3 ay önce

Tried recreating Fall Guys using the exact same prompt with Claude Opus 5.0 (Claude Code) and Kimi K3 (Kimi CLI). Kimi K3 nailed the gameplay out of the box. Physics, movement, and the overall feel were surprisingly solid Opus 5.0 had a shinier UI, but movement was broken, it kept adding unnecessary visual effects, lagged during gameplay, and spent most of its time reiterating instead of converging Runtime and cost : - Kimi K3: ~9 min, ~$4.4 - Opus 5.0: ~17 min (still iterating), ~$13+ That's roughly 2× the time and 3× the cost, yet Kimi produced the more playable result with the exact same prompt.

Tried recreating Fall Guys using the exact same prompt with Claude Opus 5.0 (Claude Code) and Kimi K3 (Kimi CLI). Kimi K3 nailed the gameplay out of the box. Physics, movement, and the overall feel were surprisingly solid Opus 5.0 had a shinier UI, but movement was broken, it kept adding unnecessary visual effects, lagged during gameplay, and spent most of its time reiterating instead of converging Runtime and cost : - Kimi K3: ~9 min, ~$4.4 - Opus 5.0: ~17 min (still iterating), ~$13+ That's roughly 2× the time and 3× the cost, yet Kimi produced the more playable result with the exact same prompt.

319,005 görüntüleme • 2 ay önce

Imagine training a model just to absolutely destroy Fruit Ninja. Peak tech

Imagine training a model just to absolutely destroy Fruit Ninja. Peak tech

755,848 görüntüleme • 10 ay önce

this is insane a high school student built a beautiful 3D interactive chemistry app

this is insane a high school student built a beautiful 3D interactive chemistry app

395,409 görüntüleme • 6 ay önce

controlling old phones with OpenClaw is insane feels like a Steve Jobs moment for AI

controlling old phones with OpenClaw is insane feels like a Steve Jobs moment for AI

283,289 görüntüleme • 8 ay önce

i used to think an AI team meant me + a few agents. then i tried Teamily AI, a Humans + Agents platform where other people sit in the same group as the agents, on the same context. different thing entirely. so: what if an AI team could find the best new generation of AI careers for you? i put 3 agents and a partner i'm working with in one group and gave them the same question: 1. AI Engineer - researched 10 genuinely new AI-native careers 2. Principal Architect - ranked the 7 most interesting ones into a page with Website Builder 3. QA Tester - opened that page, challenged the list, rejected 3 and suggested better replacements then I jumped in again with "prioritize roles with real demand and low competition", and the Architect rebuilt the page with the QA swaps and that ask. one agent researched, one ranked, one challenged the work, and a human pushed it further. one question to a finished page, all in one chat. this is what an AI team actually feels like. final page: Teamily AI try it:

i used to think an AI team meant me + a few agents. then i tried Teamily AI, a Humans + Agents platform where other people sit in the same group as the agents, on the same context. different thing entirely. so: what if an AI team could find the best new generation of AI careers for you? i put 3 agents and a partner i'm working with in one group and gave them the same question: 1. AI Engineer - researched 10 genuinely new AI-native careers 2. Principal Architect - ranked the 7 most interesting ones into a page with Website Builder 3. QA Tester - opened that page, challenged the list, rejected 3 and suggested better replacements then I jumped in again with "prioritize roles with real demand and low competition", and the Architect rebuilt the page with the QA swaps and that ask. one agent researched, one ranked, one challenged the work, and a human pushed it further. one question to a finished page, all in one chat. this is what an AI team actually feels like. final page: Teamily AI try it:

13,559 görüntüleme • 10 gün önce

KIMI K3 IS A NIGHTMARE FOR CLAUDE FABLE-5 AND GPT-5.6-SOL FOR GAME DEVELOPMENT Tried building a playable MMORPG with Kimi K3 featuring horse riding, rooftop parkour, combat and an explorable open world. The whole thing took about 55 minutes from a single prompt and cost exactly $9.37 to generate.

KIMI K3 IS A NIGHTMARE FOR CLAUDE FABLE-5 AND GPT-5.6-SOL FOR GAME DEVELOPMENT Tried building a playable MMORPG with Kimi K3 featuring horse riding, rooftop parkour, combat and an explorable open world. The whole thing took about 55 minutes from a single prompt and cost exactly $9.37 to generate.

98,360 görüntüleme • 3 ay önce

everyone wants to build a startup no one knows what to build Sam Altman explains it

everyone wants to build a startup no one knows what to build Sam Altman explains it

146,264 görüntüleme • 7 ay önce

execution beats intelligence every single time

execution beats intelligence every single time

115,800 görüntüleme • 5 ay önce

who would’ve thought i could ask an AI agent using Airtap Ai to open Zepto, build me a cart with protein cookie, drinks and groceries and it’d actually do the whole thing without me touching my phone once 😭 AI handling real mobile apps feels insane

who would’ve thought i could ask an AI agent using Airtap Ai to open Zepto, build me a cart with protein cookie, drinks and groceries and it’d actually do the whole thing without me touching my phone once 😭 AI handling real mobile apps feels insane

95,854 görüntüleme • 4 ay önce

a user had opus 4.6 thinking for 1.5 hours on a sunday non-peak hour first increase token usage, then nerf opus 4.6, and now make it buggy with nonstop feature releases what’s going on with claude code?

a user had opus 4.6 thinking for 1.5 hours on a sunday non-peak hour first increase token usage, then nerf opus 4.6, and now make it buggy with nonstop feature releases what’s going on with claude code?

118,419 görüntüleme • 6 ay önce

building side projects instead of studying cuz it’s exam season

building side projects instead of studying cuz it’s exam season

18,269 görüntüleme • 29 gün önce

average day of a remote developer

average day of a remote developer

73,176 görüntüleme • 5 ay önce

spent the day fixing tests and preparing for a new system architecture discussion + approval

spent the day fixing tests and preparing for a new system architecture discussion + approval

45,577 görüntüleme • 5 ay önce

day 1 of learning golang gonna document the entire journey through daily timelapses

day 1 of learning golang gonna document the entire journey through daily timelapses

44,184 görüntüleme • 5 ay önce

I'm selling ad space on my portfolio → 10 sponsor spots on the side rails → $60-$100/week, depending on position → visible on every page → your logo, name, copy + link your brand, on my site -

15,122 görüntüleme • 1 ay önce

bro I sent my AI agent a Google Meet link and it actually joined 😭 Pika real-time video chat for ANY agent is crazy we really out here calling our agents now

bro I sent my AI agent a Google Meet link and it actually joined 😭 Pika real-time video chat for ANY agent is crazy we really out here calling our agents now

35,303 görüntüleme • 6 ay önce

Tested out GPT-5.6 to build an AI analytics dashboard with micro animations And the fucking UI is still the same AI-generated purple slop. No massive improvements.

Tested out GPT-5.6 to build an AI analytics dashboard with micro animations And the fucking UI is still the same AI-generated purple slop. No massive improvements.

17,609 görüntüleme • 3 ay önce