See how to build AI-powered tools that run locally with Ollama. Cedric Clyburn demonstrates how to maintain data privacy, integrate Langchain, and simplify development using local AI deployment. Discover how to prototype smarter and optimize tools for enterprise tasks with ease.
I Tested Google Flow’s New AI Agent — It Changes Everything
On May 20th, 2026, Google transformed Flow from a simple AI video generator into a full-blown conversational Flow Agent — and it’s going to change how creators, filmmakers, and content studios build video projects.
In this video, I test the new Flow Agent and break down every major capability Google just added.
Here’s what we cover:
Persistent Project Memory
Flow now remembers your characters, style guides, scene rules, and creative direction across your entire project session.
Prompt-to-Edit
Edit existing footage using natural language. Color grade, speed ramp, add camera shake, adjust scenes, and make creative changes just by typing what you want.
Bulk Asset Factory
Drop in a CSV spreadsheet with dozens of prompts and let Flow generate assets simultaneously in parallel.
Flow Tools
Build custom creative tools in plain English. In the video, we create a Retro Arcade Character Card generator that turns any selfie into an 8-bit trading card.
Character Builder & Casting Studio
Create original characters from scratch, assign voice profiles, write their personality, and Flow maintains that identity across every scene.
Whether you’re a solo YouTube creator, filmmaker, game studio, or content team, Google Flow is no longer just an AI video generator — it’s becoming an AI-powered production house.
How to Generate AI Videos for FREE with ComfyUI (Step-by-Step Tutorial)
In this beginner-friendly tutorial, I’ll show you how to install ComfyUI from scratch, download a free AI video model, generate videos from text prompts, animate existing images, and save your finished videos. Everything runs locally on your computer, so there are no API keys, subscriptions, or usage credits required.
By the end of this video, you’ll know how to:
Install ComfyUI Desktop
Download the free LTX 2.3 AI video model
Generate videos from text prompts
Animate existing images
Adjust video settings like resolution, duration, and frame rate
Find and manage your generated videos
Explore additional AI video workflows
If you’re looking for a free way to create AI videos on your own PC, this tutorial will get you up and running.
Top 5 Secret A.I. Filmmaking Tools JULY 2026
Everyone is just using Seedance, but here are my secret top AI tools that changed my workflows completely… When I started a few years ago, I spent way too much money testing different tools because I wanted to understand what is good for which specific problem. And I found out: there is not one single magical tool that solves everything. Even though Seedance is really good. That’s why it’s important to have a good variety of tools in your AI toolbox. I really love every single tool I show in this video. But in the end, tools are going to stay tools. They can help us move faster and try out more ideas, but they don’t make the work matter by themselfs. What we need more than anything is soul in the work we make. A feeling, a point of view, something human that stays.
The Tools:
SwitchX (Beeble) – fixes and adapts the whole clip. Green screen, relighting, retouching, done. https://beeble.ai/
Sora – I don’t just generate with it. I pull ultra-realistic start frames and characters, then upscale. via Multimodal Platforms like Invideo, Higgsfield, Freepik etc.
Fal.ai – one dashboard, every model that exists. I use it to spot new tools before they’re famous. https://fal.ai/
Sync.so – lip-sync that keeps the actor’s real face. Fixes a line without ruining the shot. https://sync.so/
Seedaudio – changes one word in a voiceover without re-recording. Keeps the room tone intact. via Multimodal Platforms like Invideo, Higgsfield, Freepik etc.
ComfyUI Course, chapter 1 – What is ComfyUI?
Sebastian Kamph explains what ComfyUI is and how it works.
At its core, ComfyUI is a platform where users visually build AI workflows using nodes on a blank digital canvas. A node is a single functional box that performs a specific task, such as loading an AI model, receiving a text prompt, or saving the final image. By drawing lines between nodes to pass data from one to the next, users create a step-by-step workflow. These workflows can range from simple single-image generations to complex, large-scale processes that output images, videos, audio, text, or even 3D assets.
Workflows are commonly saved as standard JSON files. To share a workflow, a user simply sends this file—or even a generated output image containing the embedded metadata—and another user can easily drag and drop it directly onto their own canvas to replicate the setup.
As an open-source, entirely free-to-use program, ComfyUI runs locally on your computer. Its open-source nature allows developers globally to build thousands of “custom nodes” for highly specialized use cases, such as face-swapping or video effects. This unmatched control and customization make ComfyUI the preferred tool for professional studios.
The ULTIMATE Gemini Omni Guide: 33 Brutal Scenarios
Discover power of Google Gemini Omni Flash AI video model with these 33 ultimate generative workflows. In this detailed guide, I break down all AI video editing capabilities ranging from complex live-action VFX, smart object tracking, character replacements and much more.
The 12 Principles of Animation Explained — The Most Important Rules for Animating
In this episode of the Shot List, we break down the 12 principles of animation, how they work, and why animators continue to use them to this day.
APPLY THE 12 PRINCIPLES OF ANIMATION
Squash and Stretch: Deforming objects or characters to emphasize their weight, elasticity, or flexibility.
Anticipation: Preparing the audience for an upcoming action by showing a brief movement or change in expression.
Staging: Presenting ideas or actions in a clear and visually interesting way, directing the audience’s attention to convey the intended message.
Straight Ahead and Pose to Pose: Two different approaches to animating movement. Straight ahead animates each frame in sequence, while pose to pose plans key poses first.
Follow Through and Overlapping Action: Continuing movement after the main action stops and parts of a character or object moving at different speeds or times.
Slow In and Slow Out: Gradually increasing or decreasing the speed of movement at the beginning and end of an action.
Arcs: Animating objects and characters along curved trajectories, mimicking real-world motion.
Secondary Action: Additional movements that complement the main action, adding depth and context to a character’s performance.
Timing: The speed and rhythm of movements in animation, essential for conveying weight, energy, and emotion effectively.
Exaggeration: Pushing movements and expressions beyond reality to make them more dynamic and expressive.
Solid Drawing: Creating characters and objects with volume, weight, and dimensionality, making animation appear more believable and three-dimensional.
Appeal: Making characters and animation visually interesting and appealing, designing characters with unique personalities and maintaining consistency in design and performance.
Claude is definitely not conscious…
Anthropic just discovered a global workspace deep inside Claude’s “brain.” Are we cooked? Let’s break it down.
How To Use Google Flow For Beginners (Step-By-Step Guide)
A beginner-friendly step-by-step guide to Google Flow. How to generate images, turn them into videos, stitch clips into scenes, drop yourself in with avatars, use Agent mode, and manage your credits.
What’s at the center of Claude’s mind?
Out of everything happening in your brain right now, only a tiny fraction is consciously accessible — thoughts you can describe, hold in mind, and reason with. We found a strikingly similar divide inside our AI model, Claude.
Our experiments were inspired by a leading theory in neuroscience: the global workspace theory. It holds that a thought becomes consciously accessible when it enters a shared “workspace” that’s broadcast across the brain. We found a set of representations in Claude’s neural activity that play a similar role.