|
How AI Agents Actually Work (Every Piece Explained & Built)
Tech With Tim
AI agents are autonomous systems that use large language models (LLMs) as a reasoning core, combined with tools, memory, and planning mechanisms to complete multi-step tasks. Unlike simple chatbots, agents can decide which actions to take, call external tools or APIs, observe results, and loop back to refine their approach until a goal is achieved. The video walks through each architectural component of an AI agent — including the LLM brain, tool use, memory types, and orchestration logic — and demonstrates how to build these pieces in practice.
summary from description (full transcript skipped)
|
|
Meet Grok Bot: Your Team of AI Agents
Cursor
GrokBot er et AI-agentsystem fra xAI, der beskrives som et "team af altid aktive agenter", som brugeren kan delegere opgaver til. Hver bot har persistent hukommelse, adgang til værktøjer og sin egen computer, og de kan kommunikere med hinanden. I workshoppen demonstrerer feltingeniør Amrita, hvordan man opsætter en "stabschefsbot", der koordinerer tre underordnede bots – en indbakkeadministrator, en kalenderkoordinator og en to-do-listeorganisator. GrokBot er tilgængelig på Mac, Windows og iOS via x.ai/bot og kræver en Cursor Ultra- eller Super Grok Heavy-konto.
|
|
Skills vs MCP vs RAG vs Memory: What AI Agents Need to Know
IBM Technology
IBM Technology's video explores the four key components that AI agents rely on to function effectively: Skills, MCP (Model Context Protocol), RAG (Retrieval-Augmented Generation), and Memory. The video likely breaks down how each element serves a distinct purpose — skills enable agents to take actions, RAG provides access to external knowledge, memory allows agents to retain context over time, and MCP standardizes how agents interact with tools and data sources. Together, these components form the foundation of capable, context-aware AI agent systems.
summary from description (full transcript skipped)
|
|
Grok Bot Tutorial for Beginners (Automate Anything with AI Agents)
Zinho Automates
Zinho Automates presents a beginner-friendly tutorial on using Grok Bot to automate tasks with AI agents. The video likely covers the basics of setting up and configuring Grok's AI capabilities to build automated workflows. Viewers are guided through practical examples of how AI agents can handle repetitive or complex tasks with minimal manual input.
summary from description (full transcript skipped)
|
|
Why AI Agents Could Finally Reinvent the Credit Card
a16z
The video, from venture capital firm a16z, explores how AI agents — autonomous software systems that can act on behalf of users — could fundamentally transform the credit card industry. Without a transcript, the likely thesis is that AI agents managing purchases and financial decisions on behalf of consumers could disrupt traditional credit card models around rewards, underwriting, fraud detection, and spending authorization. The video likely argues that as AI agents become financial actors themselves, the existing credit infrastructure may need to be reimagined to accommodate machine-driven transactions rather than human ones.
summary from description (full transcript skipped)
|
|
From MLOps to AgentOps: Shipping Autonomous Agents You Can Actually Trust
Databricks Events
From MLOps to AgentOps: Shipping Autonomous Agents You Can Actually Trust — a Databricks Events presentation exploring the operational challenges and best practices for deploying autonomous AI agents in production. The talk likely covers how traditional MLOps principles must evolve to address the unique requirements of agentic systems, including reliability, observability, and trust. Key themes probably include monitoring agent behavior, managing multi-step reasoning pipelines, and ensuring agents perform predictably at scale within enterprise environments.
summary from description (full transcript skipped)
|
|
Agentic AI Program Overview
Stanford Online
Stanford Online's "Agentic AI Program Overview" introduces a program focused on agentic AI systems—AI that can autonomously plan, make decisions, and take actions to accomplish complex, multi-step goals. The program likely covers foundational concepts, tools, and frameworks for building and deploying these autonomous AI agents. It is aimed at professionals and learners looking to deepen their understanding of this emerging and rapidly evolving field.
summary from description (full transcript skipped)
|
|
Qwen 3.8 Flash Next + HERMES AGENT = AWESOME LOCAL AI AGENTS!
Digital Spaceport
Videoen viser, hvordan man kører Qwen 3.8 Flash Next-modellen i en Hermes-agent-opsætning ved hjælp af vLLM på et system med fire NVIDIA 3090 GPU'er (96 GB VRAM) og mindst 128 GB system-RAM. Værten gennemgår de tekniske konfigurationer, herunder max modellængde på 131.072 tokens, GPU-hukommelsesudnyttelse på 0,96 og aktivering af præfiks-caching og automatisk værktøjsvalg med Qwen 3 XML-parseren. Som demonstration af systemets kapaciteter viser han et webbaseret 16-bit rumvæsen-bortførelsesspil kaldet "Abductum", som agenten har genereret i blot to prompt-runder, og en komplet guide til opsætningen er tilgængelig på digitalspaceport.com.
|
|
Fal.Con 2026: Securing AI | George Kurtz, Jensen Huang, Lip-Bu Tan & Greg Brockman
CrowdStrike
CrowdStrike's Fal.Con 2026 event featured a high-profile panel discussion on securing artificial intelligence, bringing together CrowdStrike CEO George Kurtz, NVIDIA CEO Jensen Huang, Intel CEO Lip-Bu Tan, and OpenAI President Greg Brockman. The session focused on the intersection of cybersecurity and AI, addressing the growing challenges of protecting AI systems and infrastructure. The gathering of these major tech and security leaders underscores the increasing industry priority placed on AI security as adoption accelerates across enterprises.
summary from description (full transcript skipped)
|