|
Agentic AI Full Course with LangGraph: Build AI Agents & Automate Workflows (Zero to Hero) - Part 1
DSwithBappy
This is Part 1 of a zero-to-hero course on Agentic AI using LangGraph, taught by an instructor named "Bi" (DSwithBappy channel). The video outlines the full seven-phase course plan, which covers: foundational agentic AI concepts, asynchronous programming and Pydantic, LangChain basics, LangGraph core components (nodes, edges, state management, workflows), memory/planning/monitoring, and production-grade deployment with Docker, FastAPI, and AWS. The course concludes with three end-to-end real-world projects—an agentic chatbot, a custom ChatGPT-like agent, and a multi-agent travel planner called "TripMate AI." Prerequisites include advanced Python, basic generative AI knowledge, and tools such as VS Code, Git, Docker Desktop, and Postman.
|
|
This Kittl Agentic Feature Changes Everything
Philip Anders
Kittl has launched a new "Agentic AI" feature (currently in beta) that allows users to generate multiple t-shirt designs from a single simple text prompt, without needing to configure settings or write detailed prompts. The tool can create up to four design variations at once, adapt existing designs to new niches or sports by swapping text and graphics, and even generate Etsy listing mockup images by referencing an uploaded design and specifying a garment style. Users can also attach reference images to guide the AI, making it useful for brainstorming, scaling best-sellers, and producing ready-to-list product imagery with minimal effort.
|
|
AI Agents Explained: Build Your First One in 2026
Roboverse
The video explains what AI agents are and how they differ from standard chatbots like ChatGPT: rather than waiting for user input, an agent is given a goal once and then runs autonomously on a schedule, retaining memory between sessions. The host demonstrates building a personal AI agent using an open-source tool called Hermes (from Nous Research), deployed on a cloud server (VPS) via Hostinger's one-click template so it runs continuously without a local computer. The setup involves using a web terminal to configure Hermes, connecting it to AI models through Open Router (which provides access to 400+ models with a single API key), and linking it to a messaging app like Telegram for phone access. The entire process, from deployment to a working agent, takes roughly 10–15 minutes and requires no coding.
|
|
Next-Gen Self-Evolving AI Agents (ATDP)
Discover AI
This video reviews a new academic paper (dated July 2, 2026) from Ant Group and Hong Kong University of Science and Technology (HKUST) and Tsinghua University introducing a framework called ATDP (Agent Trajectory Data Protocol) for building self-evolving AI agents. The core idea is that all agent interactions—successes and failures—across any framework (LangChain, CrewAI, Hermes, etc.) should be captured as structured training signals rather than discarded, enabling continuous system improvement. The paper formally defines a self-evolving agent as a mathematical tuple encompassing the policy LLM, in-context harness, memory, tool repertoire, and guardrails, with evolution actions including LLM fine-tuning/distillation, prompt optimization, memory retrieval policy updates, and tool schema modifications. The ATDP protocol standardizes data collection across heterogeneous agent frameworks so that this holistic, interconnected optimization can be applied at both enterprise scale and by individual developers.
|
|
AI Agents For Beginners – OpenClaw Case Study
freeCodeCamp.org
This beginner-friendly course, led by Bumshad Manhatt (founder of CodeCloud), teaches AI agent development from the ground up—covering LLM fundamentals (transformers, tokenization, temperature, prompting) through multi-agent system design. Students build four hands-on agents named Zippy, Savvy, Meshy, and Cody, progressing from a single generic agent to a multi-agent orchestration setup with specialized research, memory, and coding agents. The course concludes with an in-depth case study of OpenClaw, a popular open-source AI agent, dissecting its architecture, agent loop, memory system, and security considerations. All API keys and sandbox cloud environments are provided, so learners can practice without setup hassles or unexpected charges.
|
|
MCP vs Skills: Which Is Right for Your AI Agent and LLMs?
IBM Technology
The video explains two complementary methods for adding custom capabilities to large language models (LLMs): MCP (Model Context Protocol) and Skills. MCP is a standardized protocol that connects AI agents to external data sources and services (like CRMs or databases) in a secure, permission-controlled way, abstracting APIs into an LLM-ready format—making it ideal for real-time data access. Skills, by contrast, are lightweight markdown-based files containing reusable prompts, scripts, and metadata that teach the LLM how to perform specific tasks in a consistent, repeatable manner, addressing the non-deterministic nature of LLMs. The video concludes that both tools enhance the LLM's context window and are open source and widely supported, with MCP best suited for external integrations and Skills best suited for adding domain knowledge and repeatable workflows.
|
|
How to Start an AI Agent Business as a Teenager in 2026
Jake One Page
The video presents a guide for teenagers on building an AI agent business in 2026 using a no-code platform called Base44 Super Agents. The creator argues that older approaches like freelance gigs and chatbot services are becoming obsolete, and that the real opportunity lies in deploying automated AI agents that work around the clock for multiple clients. He walks viewers through setting up a Base44 workspace — configuring the agent's memory, connecting tools like Gmail, Google Drive, and HubSpot — and demonstrates how to build a lead responder agent that automatically replies to incoming inquiries within 10 minutes and sends follow-up emails. The video also promotes a paid masterclass (offered free to viewers) covering how to build AI-powered SaaS apps, websites, and mobile apps.
|
|
Agents are slower than LLMs?
Caleb Writes Code
Agents are slower than pure LLMs primarily because they rely on external tool calls—such as web fetches, API requests, database queries, or file transfers—that can take seconds to minutes to complete, dwarfing the time the LLM itself spends generating tokens. This creates a major inefficiency at the infrastructure level, where expensive GPUs sit idle while waiting for tool results to return. To address this, techniques like disaggregated inference (separating prefill/batch processing from token generation using specialized hardware) and KV cache offloading to external storage can free up GPUs during tool execution wait times, a concept reflected in Anthropic's prompt caching feature. The video also notes that agentic workloads reshape data center resource allocation more broadly, increasing demand for CPUs, storage, and networking alongside GPUs.
|
|
Give Your AI Agent a Second Brain (Gbrain + Hermes Agent)
Tonbi's AI Garage
The video introduces Gbrain, an open-source tool (25,000+ GitHub stars) created by Y Combinator president Garry Tan, which acts as a "second brain" for AI agents like Hermes Agent by storing and searching knowledge from external sources—such as meeting notes, Obsidian vaults, and markdown files—that never passed through the chat interface. The host distinguishes Gbrain from Hermes's built-in memory (which only recalls past conversations) and LM wikis (which cover specific domains), explaining that Gbrain serves as a personal knowledge base for "your world," using hybrid vector, keyword, and knowledge-graph search exposed via 30+ MCP tools. The video walks through installation using Bun and an embeddings API key (e.g., via OpenRouter), showing how Hermes Agent can automate the setup process and then seed Gbrain with local data. A live demo shows the agent successfully retrieving meeting preparation details from ingested notes that Hermes's native memory would have been unable to find.
|
|
Tame the AI sprawl: Unify agents and MCP servers
Google Cloud Events
This Google Cloud Next session addresses the challenge of "AI sprawl" in enterprise environments, where the rapid, ungoverned proliferation of autonomous multi-agent AI systems creates serious security, compliance, and operational risks. The speakers—a Google Cloud PM, a PayPal staff engineer, and a Google Cloud principal engineer—deconstruct modern agent architecture, explaining how protocols like MCP (Model Context Protocol) and Agent-to-Agent (A2A) communication enable powerful but vulnerable multi-agent systems susceptible to prompt injection, rugpull attacks, and data exfiltration. PayPal's experience is used to illustrate real-world challenges including AI cost sprawl, governance gaps, non-deterministic outputs, and the difficulty of managing context and coordination across dozens of agents. Google Cloud's proposed solution is a "shift-down" governance platform built on five pillars—a centralized agent registry, a security gateway controlling ingress/egress, cryptographic agent identity, pluggable policy enforcement (including Model Armor), and end-to-end observability—designed to extend existing IT governance frameworks rather than replace them.
|
|
Moms Are Making $250k/Year: How To Build An Audience That Buys Using The NEW AI Agent (Step By Step)
The Calum Johnson Show
The video features an interview with John Who, founder of the creator monetization platform Stan, who claims his platform has helped 90,000 creators earn over $600 million collectively. He highlights diverse success stories, including a single mother who made $250,000 in her first year and a farm woman who earned six figures selling a sourdough bread guide, arguing that consistent daily content posting is the single most common trait among top earners. He explains that Stan's new AI tools compress the workload of content creation, making it easier for everyday people to post consistently and build profitable audiences. Beyond the financial angle, he frames entrepreneurship as a path to personal agency and self-confidence in an uncertain economic environment.
|
|
Agentic AI Full Course 2026 | Agentic AI Tutorial For Beginners | Agentic AI Course | Simplilearn
Simplilearn
Based on the title alone, this appears to be a beginner-level full course on Agentic AI published by Simplilearn, likely covering foundational concepts of autonomous AI agents, their architectures, and practical applications. The course is positioned as a 2026 tutorial targeting learners with little to no prior experience in the field. Without a transcript available, specific topics, tools, or frameworks covered in the course cannot be confirmed.
summary from description (full transcript skipped)
|