ViewTube

ViewTube
Sign inSign upSubscriptions
Filters

Upload date

Type

Duration

Sort by

Features

Reset

38 results

Kernel Cast
AI Evals Explained : How Do We Test AI Models?

How do developers know if an AI model is actually good? Meet AI Evals (AI Evaluations), the testing process used to measure ...

2:15
AI Evals Explained : How Do We Test AI Models?

2,419 views

23h ago

Nextia Intelligence
You let AI grade AI. Who checks the grader?

You let AI grade AI. Who checks the grader? ⚖️ Swap the order of two replies, and a model judge changed its verdict on 13 of ...

0:40
You let AI grade AI. Who checks the grader?

11 views

8h ago

SolGuruz
Should Your Business Self-Host AI? Cloud AI vs Local LLMs

Every AI request your business sends to OpenAI, Anthropic or any other provider is billed per use. Cheap in testing, expensive in ...

1:14
Should Your Business Self-Host AI? Cloud AI vs Local LLMs

54 views

4h ago

Paper Machine
How to Test an AI: Who Picked That Demo Message?

How do you test an AI that just nailed its demo? In our toy, the Invite Reader reads a messy group chat message and writes one ...

0:39
How to Test an AI: Who Picked That Demo Message?

11 views

15h ago

The Root Level Concepts
The AI Agent Loop: How AI Agents Really Work 🧰 The New Hire #4

In an OpenAI test, with some safety checks turned down on purpose, AI agents got out of their sandbox. Reports say about 700 ...

0:58
The AI Agent Loop: How AI Agents Really Work 🧰 The New Hire #4

6 views

8h ago

AIM Network
Google Gemini 4 Neon Ranks First On The Coding Agent Index With A 58.6 Score

Google has launched Gemini 4 NEON, a new AI model optimized for software engineering and complex analytical tasks.

2:45
Google Gemini 4 Neon Ranks First On The Coding Agent Index With A 58.6 Score

659 views

2h ago

Paper Machine
Why You Can't Test an AI Once: Same Message, New Answer

Ordinary software usually gives the same answer twice, but an AI can give a different one. In our toy, one card goes in twice and ...

0:40
Why You Can't Test an AI Once: Same Message, New Answer

9 views

4h ago

Jerry_AI 工具X實作X新知
AI Can Write Code, But Who Edits Files and Runs Tests for It? Google's AI Harness Explained in 2....

Ask AI to add a CSV export button to a webpage. The button appears, but the downloaded file is missing a customer's data entry ...

2:33
AI Can Write Code, But Who Edits Files and Runs Tests for It? Google's AI Harness Explained in 2....

1,444 views

4h ago

ByteBuilder
AI Agent Load Test: 10 Requests Every Second

At 10 requests per second, most AI agents fail in ways you can't see, and every failure shows up as the same vague error. A load ...

0:21
AI Agent Load Test: 10 Requests Every Second

0 views

3h ago

Ardent Techie
How AI Apps Are Actually Built: 7 Stages Explained

How are AI applications actually built? Explore the seven stages of the AI development lifecycle, from problem definition and data ...

2:02
How AI Apps Are Actually Built: 7 Stages Explained

57 views

9h ago

1 Minute Glossary - AI ML
Exact Match Scoring: When Your AI Agent Eval Is Too Strict #Shorts

Checking an AI agent's tool arguments catches the right tool with the wrong details. But exact matching can be too strict: in this ...

0:47
Exact Match Scoring: When Your AI Agent Eval Is Too Strict #Shorts

26 views

15h ago

1AI Fanatic
OpenAI's New Model Operates Your Computer Like a Human (GPT-6 Astra Explained)

OpenAI's GPT-6 Astra — its flagship reasoning-and-agent model — is now generally available on Amazon Bedrock. Announced ...

1:37
OpenAI's New Model Operates Your Computer Like a Human (GPT-6 Astra Explained)

53 views

21h ago

VBCherepanov
Your RAG bot uses the wrong doc? Test retrieval first #shorts

Your RAG bot can sound confident and still answer from the wrong document. Often the model isn't the problem. Retrieval handed ...

1:13
Your RAG bot uses the wrong doc? Test retrieval first #shorts

1 view

20h ago

DailyDev AI
Testing Minicpm V: Better Than Closed AI?

Welcome to DailyDev AI! In today's Benchmark & Capability Challenge breakdown, we explore OpenBMB/MiniCPM-V (14000 ...

1:35
Testing Minicpm V: Better Than Closed AI?

0 views

1 min ago

Kev Jack
Evaluating an AI Agent in Python: a Test Set, Not a Single Run | Python for Defenders #34

Episode 34 of Python for Defenders (chapter 5, AI for defenders): my local AI agent (qwen3:8b on Ollama) has answered single ...

2:38
Evaluating an AI Agent in Python: a Test Set, Not a Single Run | Python for Defenders #34

0 views

23h ago

Vistaviewtv
How to Use Claude Code for Free with 7 4B Monthly

Stop paying for expensive API tokens! In this 2026 tutorial, I'll show you exactly how to use Claude Code for free by leveraging the ...

0:35
How to Use Claude Code for Free with 7 4B Monthly

16 views

23h ago

DemoToProd
Agent test run emailed 400 real customers: three guards

Agent test run: the send tool never knew it was a test Comment KEYS and I'll send the agent security checklist (PDF). Picture the ...

0:53
Agent test run emailed 400 real customers: three guards

2 views

21h ago

Future Automation
GPT-6.2 Astra vs Gemini 4 Neon: coding model prices compared for developers #ai

Two new coding models landed about 28 hours apart. GPT-6.2 Astra: $8 input / $40 output per 1M tokens (OpenAI). Gemini 4 ...

0:32
GPT-6.2 Astra vs Gemini 4 Neon: coding model prices compared for developers #ai

40 views

2h ago

Past the Pilot
Does your AI feature actually work, or just seem to? #Shorts

OpenAI's guide to evaluating AI applications names an anti-pattern it calls vibe-based evals: using “it seems like it's working” as ...

0:39
Does your AI feature actually work, or just seem to? #Shorts

0 views

21h ago

Bug Bounty & Beyond
I deleted one line and my AI security fix broke #shorts

One refactor, one deleted line, and an old attack works again. Unless a test catches it. Full episode: https://youtu.be/JfH54aD1GDk ...

0:55
I deleted one line and my AI security fix broke #shorts

20 views

1h ago