
Aakash Kanojiya
@Kanojiyaaakash1 • 40,237 subscribers
I explain AI and tech tools for builders (200K+ of you) | Founder | ex Full Stack Developer | ✉️ [email protected]
Shorts
Videos

🚨 BREAKING: Big day for voice AI. Cartesia just dropped Sonic 3.5 and Sarvam announced their $300M Series B - so I decided to put both to the test myself. Same scripts, side by side. The results are kind of insane. Sonic came out a leap ahead on emotion, alphanumerics, and switching across Indian languages mid-sentence. Karan Goel grew up in Delhi, went to IIT Delhi, and constantly talks about how much he cares about building for India. Hard to argue he doesn't mean it 👇
Aakash Kanojiya708,039 просмотров • 4 месяцев назад

A stealth model with 1M context just dropped. It’s called Space Bunny. → 1M context → Multimodal → Zero Data Retention So naturally, I gave it one job: Build a GTA style quiz Game from scratch. No starter code. No assets. Just one prompt and Space Bunny doing everything. It’s designed to handle long, complex coding tasks while keeping track of the broader codebase. Let’s see how far 1M context can take an agent without losing track of the codebase.
Aakash Kanojiya74,112 просмотров • 26 дней назад

I asked Codex with GPT Astra 6 to build me a 3D flight game. I could’ve built the dragon manually. Instead, I connected Hyper3D Rodin via MCP to generate the complex model faster and with better quality Minutes later, I had a fully playable 3D game running in my browser. Here’s the workflow ↓
Aakash Kanojiya76,805 просмотров • 1 месяц назад

How well can Sarvam say a Hindi tongue twister? I did a comparison with Cartesia
Aakash Kanojiya89,048 просмотров • 1 месяц назад

I gave Cline Desktop app a repo with 3 intentional vulnerabilities. Hardcoded password. SQL injection. Shell injection. I gave it a cron job “Scan this codebase every night. Flag the issue, severity, risk, and fix.” At 2 AM, the agent ran the scan while I was asleep. All 3 vulnerabilities were caught. But the interesting part came next. → Asked it to fix the SQL injection with parameterized queries → Reverted to the pre-fix checkpoint → Forked the session → Asked for the most conservative fix possible And it all runs inside Cline Desktop - an open-source app for open-weight models. No editor open. No manual trigger. No babysitting. Just a scheduled agent working inside a real workspace, using the model I chose. This is the shift: AI coding agents aren't just tools you open anymore. They're starting to feel like teammates that work while you're offline.
Aakash Kanojiya25,827 просмотров • 1 месяц назад
Больше нет контента для загрузки