Prompt Engineering - Videos
Back to Channel
OpenAI Codex App: Claude Cowork Killer?
Checkout the next codex app from openai: https://openai.com/index/introducing-the-codex-app/ My Dictation App: www.whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https:/...
Why Agents Are Ignoring Your Skills (Literally)
Vercel's latest evaluations reveal that AI agents ignore 'Skills' over 56% of the time, casting doubt on the effectiveness of progressive disclosure for current models. I break down why simple "con...
Stop Typing. (Seriously)
Get the App here: https://engineerly76.gumroad.com/l/pbzck Use code PROMPT for 25% off! I built the ultimate local-only voice transcription tool for macOS that lets you code and write faster than...
Kimi K2.5 - Its more than an LLM
Kimi K2.5 is here. Its a multimodal model specifically trained for agentic use. https://x.com/Kimi_Moonshot/status/2016019760308457649 https://www.kimi.com/blog/kimi-k2-5.html https://www.kimi.co...
Clawdbot (moltbot) Clearly Explained...
Clawdbot is blowing up on X/Twitter. Here is my take on it. https://clawd.bot/ My voice to text App: whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-sit...
Stop Using AI Agents Like Chatbots. (Do This Instead)
Checkout Verdent here: https://www.verdent.ai/?id=700231 use my personal code: 4S62V4 for 60% off Stop treating AI agents like chatbots—the secret to 'vibe coding' isn't the model, but the paralle...
Qwen3-TTS: The ElevenLabs Killer?
Qwen3-TTS is a groundbreaking open-source model that brings real-time voice cloning and custom voice design directly to your local machine. I breakdown the 1.7B and 0.6B models, demonstrating how t...
"Don’t Outsource Your Thinking" to your Agent
AI coding agents can drastically speed up your workflow, but without the right habits, they often lead to "context rot" and unmaintainable code. This video breaks down 10 essential lessons from a y...
Amazon’s New AI Agent is... Boring? (And that’s good)
Get started with Amazon Nova Act for free here: https://fandf.co/4rMkZYd Browser automation often breaks when you move from a demo to the real world. In this video, we explore Amazon Nova Act appr...
NVIDIA Nemotron ASR... The Whisper Killer?
NVIDIA’s new Nemotron Speech ASR uses cache-aware streaming to eliminate the latency drift found in sliding window models like Whisper. This video explains the architecture changes and demonstrates...
DeepSeek Engram: We’ve Been Building LLMs Wrong
In this video, I delve into a groundbreaking paper by DeepSeek called Engram that addresses the inefficiencies of transformer-based large language models (LLMs). Traditional LLMs use deep computati...
Anthropic Just Fixed MCP’s Biggest Problem
In this video, I discuss how Anthropic has addressed the issue of context window pollution caused by MCP server tool definitions. By using the new MCP tool search, you can dynamically load tools wh...
Claude Cowork: First General Agents from Anthropic
In this video I dive into @anthropic-ai new "Cowork" agent, a desktop application that gives AI the power to access your local files and control your browser to automate complex workflows. In this...
RAG is Dead? Try Agentic File Search
In this video we will look at file search exploration as a potential replacement to RAG. This is built on top of the the fs-explorer from llamaIndex. LINK to the Repo: https://github.com/PromtEng...
The Holy Grail of Intelligence - Explained.
I break down the "5 Levels of Continual Learning" to explain why current AI models suffer from catastrophic forgetting and remain frozen after training. We also explore whether engineering hacks li...
Ralph Wiggum - Ship Code While you Sleep!
This video explains the 'Ralph Wiggum' pattern—a method for running Claude Code autonomously using a simple bash loop. I cover the theoretical concept, when to use this approach versus standard pro...
Claude’s Best Hidden Feature
In this video we look at Anthropic's "Ask User Question Tool" for specs driven development. It interviews the user for collaboratively working on the SPECS for a project. My voice to text App: wh...
Open Source AI Voice Is Finally Good!
In this video, I test Chatterbox Turbo by Resemble AI. It's a permissive, MIT-licensed model that supports zero-shot voice cloning, multilingual output, and paralinguistic control (laughs, pauses, ...
""I have never felt this much behind as a programmer" - Andrej Karpathy
LINKS: https://x.com/karpathy/status/2004607146781278521 My voice to text App: whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-site.thinkific.com/courses/...
NVIDIA DGX Spark: Unboxing and Getting Started
The NVIDIA DGX Spark is a dedicated local AI workstation powered by the Grace Blackwell superchip and 128GB of unified memory, designed to let developers build, train, and run massive models withou...
The New Gemini Text to Speech
Checkout Blink.New for building fullstack applications! https://blink.1stcollab.com/engineerprompt We look at the new Gemini text to speech built on top of Gemini 2.5 pro and flash. You can contro...
Gemini 3 Flash is here ...
Google DeepMind just released Gemini 3 Flash, and it’s officially their biggest update to the Flash series yet. This model is not just faster—it actually outperforms Gemini 3 Pro on key coding benc...
GPT 5.2 is here.. the biggest update yet?
OpenAI just released GPT-5.2. This is the biggest update to the GPT-5 family. My voice to text App: whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-site....
Mistral Devstral 2: The DeepSeek Killer we wanted?
Devstral 2 is the latest coding model from Mistral in 2 flavors. My voice to text App: whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-site.thinkific.com...
DeepSeek Speciale: How They Did It Again!
The Whale is back with a huge release - 2 new models, Deepseek v3.2 and Deepseek v3.2-Speciale My voice to text App: whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https:...
Meet Junie: JetBrains' New Autonomous Coding Agent.
Checkout Junie for JetBrains' IDEs: https://jb.gg/Junie-Engineer Junie takes a completely different approach to coding assistance by planning, executing, and verifying each change instead of blind...
Meta is Back! Segment Anything 3 is Here (Open Weight)
Meta release segment anything model 3 along with SAM 3D, a unified model for detection, segmentation, and tracking of objects in images and video using text, exemplar, and visual prompts. https://...
Google’s nano banana pro just killed Photoshop...
the new nano banana pro is here. Its build on Gemini 3 pro. aistudio.google.com My voice to text App: whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-si...
Emergent: The First REAL AI Full-Stack Developer?
Check out Emergent for full stack development: https://app.emergent.sh/?via=engineerprompt In this video, I put Emergent to the test of replicating my RAG as a service platform. Its not just anoth...
Google Antigravity - Did Google Just Killed Cursor?
Google just released Antigravity, a new type of coding agent that is your junior Dev. https://blog.google/products/gemini/gemini-3/ https://blog.google/technology/developers/gemini-3-developers/ ...
Gemini 3 Pro: It was worth the wait!!!
Gemini 3 pro is here and its the most capable Gemini model to day. Its built for agentic era. We will have a very first look at the model in this video. we will also have a look at Google antigrav...
Google File Search API - Managed RAG for Everyone
Thanks to clerk for making this video possible. Check them out here: https://go.clerk.com/YZA4lQK Google just launched Gemini API File Search—a fully managed RAG you can use with one API call. I d...
This is the fastest voice to text and speaker diarization App
Get the fastest Voice-to-Text App here: https://whryte.com/ Feature request and feedback: https://tally.so/r/wo4ky1 Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-...
Is This the End of MCP for AI Agents?
MCP (Model Context Protocol) sounds great, but in practice it bloats an AI agent’s context with tool definitions and intermediate results—burning tokens and adding noise. This video shows why code-...
No More IDEs - Is this the End?
Interested in securing the AI generated code? checkout Snyk: https://snyk.plug.dev/a2h64Ve Learn why model makers are shipping terminal-first coding agents—from Claude Code to Qwen/Kimi/Minimax M2...
Google Goes All-In on Vibe Coding with AI Studio
Checkout Clerk for your Authentication and User Management needs: https://go.clerk.com/ED6gexZ Google goes all in on vibe coding with their new AI Studio platform. aistudio.google.com Ge my lo...
Llama.cpp Gets a New Web UI
Learn how to get started with llama.cpp new webui. LINKS: https://github.com/ggml-org/llama.cpp/discussions/16938 https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md Ge my local Vo...
What Makes Qwen 3 Max Thinking So Weird?
In this video, I have a look at Qwen -3 Max Thinking. Test it out here: https://chat.qwen.ai/ Ge my local Voice-to-text app: https://whryte.com/ Website: https://engineerprompt.ai/ RAG Beyond Bas...
This Open AI model is a Coding Beast
We are looking at MiniMax M2, the best open weight model! LINKS: https://www.minimax.io/news/minimax-m2 https://agent.minimax.io/ https://zswn3lcow8wq.space.minimax.io/ https://s2b6x2wq9w3s.space...
Claude Desktop Is Now Even Better — Talk to Claude, Code in Browser, and More!
Check out Kimi Ok Computer here: https://www.kimi.com/?utm_campaign=TR_rhliwcOs&utm_content=&utm_medium=Youtube&utm_source=CH_kEMBez3l&utm_term= Anthropic just dropped major updates for Claude — i...
Build Custom Skills for Claude Agents
Checkout Clerk for your Authentication and User Management needs: https://go.clerk.com/di1ufSF In this video, I show how I give Claude new coding skills by feeding it real documentation through An...
OpenAI Joins the Browser War with ChatGPT Atlas
OpenAI just launched ChatGPT Atlas. It's first-ever browser built around AI and memory, and it might be the biggest threat yet to Google Chrome's dominance. In this video, I test Atlas hands-on, co...
The Ultra-Fast and Local Speech Dictation System
About one of the fastest real-time on-device speech-to-text apps for macOS. No internet, no data sharing; just instance transcription. In this video, I'll show you how it works. It's also powered b...
Claude Skills: Glimpse of Continual Learning?
In this video we look at Claude Skills and how they are different than MCP, Sub-agents and slach commands. You will learn how skills enable modular, efficient workflows with domain knowlege. This b...
Why they feel dumber? ... its not the model
Thanks to Supabase for sponsoring this video, check them out: https://supabase.plug.dev/UGNnsbw Learn why the same 1T open-weight model can perform widely differently, depending on the inference p...
Haiku 4.5 Is Here—And It’s a Beast at Coding
A quick look at the Veo 3.1 and Haiku release. https://www.anthropic.com/news/claude-haiku-4-5 https://developers.googleblog.com/en/introducing-veo-3-1-and-new-creative-capabilities-in-the-gemini...
Vibe for Wordpress - 10Web new AI Builder (no code)
Check out 10web for building and hosting wordpress websites: https://10web.io/?utm_source=YouTube+&utm_medium=Influencer&utm_campaign=engineerprompt I tested 10Web, the first all-in-one AI platfor...
3 New Context Engineering Skills for Agents
Try Skywork ai at: https://skywork.ai/p/O1h1m4 use the code: engineerprompt for discounts. It has support for full stack app development with Supabase integration. In this video we will explore...
Ling 1T Model: Too Good to Be True?
Access Ling 1T: https://huggingface.co/inclusionAI/Ling-1T https://x.com/AntLingAGI/status/1975942293330018426 https://zenmux.ai/inclusionai/ling-1t?utm_source=hf_inclusionAI Let's look at Ling-1T...
OpenAI Just Confused Everyone... Again
OpenAI’s new Agent Kit has everyone confused! is it really building agents or just fancy workflows? In this video, I break down what OpenAI means by “agents,” how it differs from Anthropic’s defini...