Nainsi Dwivedi
@NainsiDwiv50980 • 17,786 subscribers
I don’t code. I build leverage with AI. AI tools • vibe coding • SaaS growth Helping builders turn ideas → products → traction - DM for collabs 📩
Shorts
Videos
This is wild 😱 Anthropic just made a huge mistake. They pulled Claude Code from the $20 plan… So the internet did what it always does: Rebuilt it. Better. Open-source. Introducing OpenClaude ↓ • Pick ANY model (GPT, Gemini, DeepSeek, local) • Run it locally or in the cloud • No subscriptions • No limits • Full coding agent in your terminal This isn’t just a clone. It’s more flexible than Claude Code ever was: → Multi-provider support → Tool-driven workflows (bash, files, agents, MCP) → Streaming + real-time execution → Works with Ollama, OpenAI, GitHub Models & more And the craziest part? You can route different agents to different models for cost + performance optimization. Frontend → cheap model Planning → powerful model One CLI. Infinite setups. This is what happens when you try to gatekeep dev tools. The community builds something unstoppable. OpenClaude is going to explode.
Nainsi Dwivedi43,438 views • 1 month ago
NVIDIA just gave every robotics beginner a cheat code. You don’t need expensive hardware. You don’t need to burn motors. You don’t need to break prototypes. You can skip all of that. Because you can build the entire robot… virtually. With Isaac Sim, you design everything in simulation: Chassis. Wheels. Joints. All running on real physics. You add sensors—RGB cameras, 2D LiDAR—so your robot can actually see. You wire up control systems using ROS 2 and OmniGraph. You stream live sensor data straight into RViz. And suddenly… You’re not guessing anymore. You’re thinking like a robot. That’s the real shift. Simulation isn’t just practice. It’s leverage. You iterate faster. You break nothing. You ship something that actually works—before touching the real world. If you’re trying to get into robotics or physical AI… This isn’t optional. This is the new starting line.
Nainsi Dwivedi26,086 views • 1 month ago
Claude Code made developers 10x faster. Git worktrees just broke the speed limit again. Teams are now running 4–8 AI agents in parallel — each in its own branch — working at the same time. Not sequential. Not one prompt at a time. Parallel AI engineering. Here’s the shift: Old workflow: Prompt → wait → review → fix → repeat New workflow: Agent 1 → builds feature Agent 2 → writes tests Agent 3 → reviews security Agent 4 → refactors architecture Agent 5 → updates docs All running simultaneously. This is called compound engineering. You’re not using AI as a tool anymore — you’re managing a team of agents. Git worktrees make this possible: Each agent gets: • its own branch • its own workspace • its own context • its own task No conflicts. No overwriting. No chaos. Just parallel progress. The result? Claude makes you 10x faster. Parallel agents multiply that again. 10x → 30x → 50x productivity. Senior engineers are already doing this: brainstorm agent planning agent implementation agent review agent testing agent It’s basically CI/CD for thinking. And the weirdest part: This actually slows each step down but speeds the entire system up. More planning More reviewing More parallel execution = higher quality + faster delivery. We’re moving from: "prompt engineering" to: "agent orchestration" The new skill isn’t coding. It’s managing AI workers. The best engineers won’t write more code — they’ll run more agents. Bookmark this. Parallel AI development is the next meta
Nainsi Dwivedi33,137 views • 1 month ago
Holy shit... One month before Andrej Karpathy dropped the “LLM Wiki” idea… someone had already built it. Not theory. Not hype. A working system. LLMs today restart from zero every time. No memory. No evolution. No learning. That’s broken. Most people use CLAUDE.md / AGENTS.md → Manual updates → Easy to forget → Knowledge gets lost That’s not intelligence. That’s a notepad. So I built dynamic memory for LLMs: • Learns automatically • Stores without manual effort • Weakens irrelevant data over time • Strengthens what actually matters Before the “LLM Wiki” idea went viral, this already had: • Brain-like decay/boost memory • Smart deduplication (no noise) • Experience tracking per tech • Cross-referenced knowledge graph • Loss-proof backups • Auto-grouped tool sessions Then I added: • memory_audit (clean + optimize memory) • Memory injection directly into tools Now this isn’t just a concept. It’s a live system that turns LLMs from stateless tools → evolving intelligence. Using it daily: → ~5K tokens → Zero manual work → Memory that actually compounds Link in comments. RT & Bookmark — this will be standard soon.
Nainsi Dwivedi15,726 views • 1 month ago
Stop...Your AI shouldn’t sit in a sidebar. It should actually work inside your notes. AnySlate just made that real. Your AI can: • read your markdown files • edit sections • create docs • search your workspace • organize everything No copy paste No switching tabs No proprietary lock-in Every file is a real .md file Your notes stay portable forever And the wild part — it supports MCP natively So Claude, Cursor, or Windsurf can directly work inside your workspace This is the shift from AI that helps you write to AI that manages your knowledge Real-time collaboration Built-in AI assistant PDF + web publishing Version history Works on Mac Windows Linux + browser This is what a modern markdown editor should look like
Nainsi Dwivedi13,617 views • 2 months ago
No more content to load