Invidious
Log in
Zero to MVP
Subscribe | 33K
 RSS

AI that actually works — tested on real tasks, with real numbers.

I'm Nick Blokhin, a software architect with 20+ years of shipping production systems. I test AI models and tools the way you'd actually use them — on real hardware, on real work — and tell you what's worth your time and money, and where it breaks.

A lot of that is AI you run yourself: local models on laptops, mini PCs, gaming GPUs, even a Raspberry Pi. The rest is the models and tools everyone's talking about, tested the same way, without the hype.

What you'll find here:
• Which AI models are actually worth using — local and cloud
• AI on your own hardware: what fits, how fast, what it costs
• AI coding tools in real work: Claude Code, VS Code, Zed
• Honest verdicts: can a free model replace a paid subscription?

One AI signal, one tool, one MVP idea — a practical 5-minute email every week → weekly.blokhin.us

View channel on YouTube
Switch Invidious Instance
Videos
Playlists
Posts
newest
oldest
popular
Next page  

6:15

Ripple 65B: The New Best US Open Coding Model?

Zero to MVP

Shared 1 day ago

7.2K views

8:35

15 Cents a Run: Are Chinese Flash Models Good Enough to Code?

Zero to MVP

Shared 4 days ago

11K views

8:01

I Treated AI Video Like Code. Here’s What Broke

Zero to MVP

Shared 1 week ago

23K views

4:09

How My MacBook Turns Into an AI Workstation

Zero to MVP

Shared 1 week ago

22K views

8:10

$7 vs $0.65: Opus 5.5 and MiniMax M3.1 Flash on My Go Project

Zero to MVP

Shared 1 week ago

10K views

5:55

Kimi K3.1 Pro Is Nearly Opus 5.5 on Paper. What About a Real Project?

Zero to MVP

Shared 2 weeks ago

11K views

7:29

Claude vs DeepSeek vs Local Qwen 3.8: Who Directs a Better AI Video Edit?

Zero to MVP

Shared 2 weeks ago

17K views

8:58

Can an Agent Make a Small Local Model Reliable?

Zero to MVP

Shared 2 weeks ago

21K views

6:01

New DeepSeek V4.1 Flash vs Claude Opus 5.5 on a 20,000-Line Project

Zero to MVP

Shared 2 weeks ago

58K views

7:08

I Forced AI to Promote My App (And It Invented Features)

Zero to MVP

Shared 3 weeks ago

26K views

6:06

Can a 1B AI Model Actually See? I Tested 29 Tasks

Zero to MVP

Shared 3 weeks ago

24K views

8:13

Claude Haiku 5.5 vs Local Qwen 3.8 27B: Is Local AI Still Worth It?

Zero to MVP

Shared 3 weeks ago

84K views

7:14

A 125B Local Model on a 16GB GPU: Fast, But Is It Smart?

Zero to MVP

Shared 1 month ago

73K views

7:03

No Designer, One Prompt: Opus 5.5 Made My Launch Page and Promo Video

Zero to MVP

Shared 1 month ago

6.1K views

10:03

Which Local AI Model Should You Run on a Raspberry Pi 5?

Zero to MVP

Shared 1 month ago

44K views

9:44

GPT-6 Sol vs Opus 5.5 — Picking My New Daily Driver

Zero to MVP

Shared 1 month ago

7.3K views

9:14

I Built a Scheduler So My Local LLMs Can Share One GPU

Zero to MVP

Shared 1 month ago

13K views

12:23

The Best Local LLMs You Can Run on a Normal Home Computer

Zero to MVP

Shared 1 month ago

124K views

2:07

Ornith 1.5 Speed on a MacBook M4 Pro — 9B and 35B

Zero to MVP

Shared 2 months ago

15K views

10:31

Qwen 3.8 Locally: How Much Can You Actually Trust It?

Zero to MVP

Shared 2 months ago

150K views

9:05

Why I Run 2–4 GB AI Models 24/7

Zero to MVP

Shared 2 months ago

493K views

10:16

Gemma 4 12B Coder Tested: Where Small Local Models Break

Zero to MVP

Shared 3 months ago

49K views

6:43

Testing Qwen 3.8 on a Real App, Not Benchmarks

Zero to MVP

Shared 3 months ago

7.8K views

9:28

Can a Small Local LLM Actually Build a Useful App?

Zero to MVP

Shared 3 months ago

18K views

7:15

GLM 5.2 vs Kimi K2.7 Code — Which Is Better for Coding?

Zero to MVP

Shared 4 months ago

15K views

6:39

Testing GLM 5.2 on Easy, Medium, and Hard Coding Tasks

Zero to MVP

Shared 4 months ago

21K views

6:34

Getting Started with Hermes Agent — Install, Configure, Run

Zero to MVP

Shared 4 months ago

7.1K views

3:20

Can Hermes Agent Use Claude Code? (Yes — Here's How)

Zero to MVP

Shared 5 months ago

38K views

12:35

Hermes + DeepSeek 4 + MiniMax 2.7: Multi-Model Coding on a ZimaBoard

Zero to MVP

Shared 5 months ago

18K views

6:07

Gemini 3.5 Flash First Look — Real Coding Test

Zero to MVP

Shared 5 months ago

12K views

3:02

From Gemma 4 Test Task to Real App: Introducing Texodus

Zero to MVP

Shared 5 months ago

5.5K views

10:35

Qwen 3.6 vs Gemma 4: I Built the Same App With Both Locally

Zero to MVP

Shared 5 months ago

275K views

9:19

Testing Qwen 3.6 Locally on Easy, Medium, and Hard Coding Tasks

Zero to MVP

Shared 6 months ago

163K views

6:26

Zed Editor + Local LLMs: Full Setup Guide with Ollama and LM Studio

Zero to MVP

Shared 6 months ago

26K views

3:25

How to Connect DeepSeek v4 to VS Code in 2 Minutes

Zero to MVP

Shared 6 months ago

79K views

8:40

DeepSeek v4 + Hermes Agent + ZimaBoard 2: My New AI Coding Setup

Zero to MVP

Shared 6 months ago

38K views

8:46

Hermes Agent on Raspberry Pi: Self-Improving AI in Under 8 Minutes

Zero to MVP

Shared 6 months ago

56K views

3:19

Claude Opus 4.7 Built My Desktop App From an Empty Folder

Zero to MVP

Shared 6 months ago

4.1K views

6:30

Claude Code + VS Code + Local LLM: The Perfect Dev Setup

Zero to MVP

Shared 6 months ago

38K views

8:21

Coding in VS Code with Gemma 4 and Ollama

Zero to MVP

Shared 6 months ago

203K views

10:03

How I Use Claude Code with Gemma 4 (Local LLMs, No API Costs)

Zero to MVP

Shared 6 months ago

219K views

9:09

Gemma 4 on Raspberry Pi 5: A Surprisingly Usable Local AI Setup

Zero to MVP

Shared 7 months ago

422K views

8:14

Gemma 4 - I Tested it on My Laptop and Desktop

Zero to MVP

Shared 7 months ago

219K views

2:14

I Ran a 48GB AI Coding Model Locally — Here's How Fast It Actually Is

Zero to MVP

Shared 7 months ago

5.3K views

8:19

Nemotron 3 Super: Can an 83GB Model Run on 16GB VRAM?

Zero to MVP

Shared 7 months ago

23K views

8:31

Qwen Coder Next Locally: Can It Replace Paid AI Models?

Zero to MVP

Shared 7 months ago

105K views

2:14

Rust vs Nim vs C: How Big Is Your Hello World?

Zero to MVP

Shared 7 months ago

5.3K views

9:19

Qwen 3.5 Locally: Can It Replace Paid AI Models?

Zero to MVP

Shared 7 months ago

175K views

4:51

Password Manager Devlog #2: Database Decryption, Entries & Saving

Zero to MVP

Shared 7 months ago

993 views

8:23

Why I Decided to Build My Own Password Manager

Zero to MVP

Shared 8 months ago

1.8K views

9:27

Choosing a New Color Theme for Zed Editor

Zero to MVP

Shared 8 months ago

6.4K views

5:46

Run AI Models Locally for FREE - Simple 3-Step Guide (Ollama Tutorial)

Zero to MVP

Shared 9 months ago

2.6K views

10:03

7 MCP Servers That Turn AI Into a Real Developer Assistant

Zero to MVP

Shared 9 months ago

3.1K views

8:40

What is MCP? (And Why You Must Start Using It)

Zero to MVP

Shared 9 months ago

878 views

3:33

3 Ways to Backup All Your GitHub Repositories

Zero to MVP

Shared 10 months ago

1.8K views

5:31

I Tested Google Antigravity’s Browser Control — How It Works

Zero to MVP

Shared 10 months ago

10K views

4:36

Google Antigravity: Can It Replace VS Code?

Zero to MVP

Shared 10 months ago

26K views

8:57

Google Antigravity Explained: Complete Beginner Tutorial to Get Started Fast

Zero to MVP

Shared 10 months ago

1.6K views

6:00

Zed Live Server: The BEST Way to Run Live Reload in Zed

Zero to MVP

Shared 11 months ago

4.3K views

8:54

LM Studio Tutorial & Review – The Best Way to Run AI Locally

Zero to MVP

Shared 11 months ago

30K views

Next page