סוכני AI

תוכן הקורס

1. 1CrewAI - Your Own Team of Autonomous Agents10:32
2. 2AutoGen Studio with 100% Local LLMs (LM Studio)9:46
3. 37 Custom GPTs to Try for Productivity and Fun11:27
4. 4Claude 3 Release and The Problem with Benchmarks17:19
5. 5Introducing DEVIKA - OpenSource AI Software Engineer | Local Install15:53
6. 6SWE-Agent: The New Open Source Software Engineering Agent Takes on DEVIN7:55
7. 7Cohere's Command-R+ Specialized Model for RAG and Tools9:23
8. 8Claude 3 Introduces Function Calling and Tool Usage14:15
9. 9Taking Function Calling to the NEXT Level with Groq API 🚀 🚀 🚀10:54
10. 10Agents with Vision 👀 - Powered by Claude 3 Haiku & Opus13:18
11. 11MIXTRAL 8x22B: The BEST MoE Just got Better | RAG and Function Calling10:53
12. 12How Good is Phi-3-Mini for RAG, Routing, Agents21:07
13. 13Creating J.A.R.V.I.S.2:35
14. 14Creating JARVIS - Your Voice Assistant with Memory13:47
15. 15Gemini Flash is SURPRISINGLY Good for Agents and Function Calling18:30
16. 16Qwen-Agent: Build Autonomous Agents with The Best Open Weight Model19:34
17. 17Claude Artifacts: What it can do and limitations12:37
18. 18Introducing Gemini API Code Interpreter (Execution)18:21
19. 19Making Long Context LLMs Usable with Context Caching13:39
20. 20Graph RAG: Improving RAG with Knowledge Graphs15:58
21. 21Graph RAG with Ollama - Save $$$ with Local LLMs12:09
22. 22Multi-modal RAG: Chat with Docs containing Images17:40
23. 23Agentic RAG: Make Chatting with Docs Smarter16:11
24. 24ColPali: Vision Language Models for Efficient Document Retrieval17:36
25. 25GPT-4o mini - Can it be used for Agentic RAG?12:51
26. 26Just Large Enough - Mistral released a BEAST17:38
27. 27Create Anything with LLAMA 3.1 Agents - Powered by Groq API22:54
28. 28Gemini 1.5 Pro Experimental - New Features with Major Upgrade15:24
29. 29Custom Chatbots for your Business Needs - No Code Platform11:43
30. 30Gemini Agents are Surprisingly Good18:31
31. 31Agents, Agents and more Agents 🤖🤖🤖14:09
32. 32Not Just Talk: A Voice Assistant That can take Actions14:35
33. 33Anthropic’s Claude System Prompt Revealed: Key Takeaways for Developers!18:52
34. 34Run Uncensored LLAMA on Cloud GPU for Blazing Fast Inference ⚡️⚡️⚡️13:42
35. 35OpenAI O1 Can Reason—But How Good Is It?16:39
36. 36O1’s Chain of Thought: I Built a System to Mimic It—Here’s How It Went!16:29
37. 37EASIEST Way to Fine-Tune LLAMA-3.2 and Run it in Ollama17:36
38. 38OpenAI Canvas First Impressions: Will It Replace Your Coding Assistant?10:56
39. 39Goodbye Text-Based RAG, Hello Vision AI: Introducing LocalGPT Vision!17:58
40. 40Stop Losing Context! How Late Chunking Can Enhance Your Retrieval Systems16:49
41. 41How OpenAI’s SWARM Simplifies Multi-Agent Systems10:29
42. 42LightRAG: A More Efficient Solution than GraphRAG for RAG Systems?19:49
43. 43Meta New Models - SAM 2.1, Spirit LM, MEXMA and More10:22
44. 44Local LightRAG: A GraphRAG Alternative but Fully Local with Ollama18:55
45. 45Claude 3.5 Sonnet Upgrade + New API to Control Your Computer – First Look8:55
46. 46Ollama with Vision - Enabling Multimodal RAG13:01
47. 47Do Anything with Local Agents with AnythingLLM10:45
48. 48Gemini 2.0 Flash in Action: How Multi-Modal AI is Changing Everything18:02
49. 49Gemini 2.0 Flash Thinking - Does it Pass the Misguided Attention Test?14:23
50. 50Meet KAG: Redefining RAG Systems with Advanced Reasoning14:18
51. 51DeepSeek v3: The First Open Model to Rival OpenAI and Anthropic!27:45
52. 52Anthropic’s Blueprint for Building Lean, Powerful AI Agents28:25
53. 53SmolAgents: A Smol Library to Build Agents24:33
54. 54Build Multi-Agent Systems with SmolAgents25:53
55. 55Google's Blueprint to Building Powerful Agents17:31
56. 56Easiest Way to run Private RAG on Windows5:56
57. 57ChatGPT Tasks: Here’s How to Use Them Right8:46
58. 58EASIEST Way to Train Custom Reasoning (CoT) Model - From Data Prep to Inference18:58
59. 59Devin AI - Is the First AI Software Engineer Ready? Real World Testing Results14:13
60. 60Deepseek R1 [Tested]: Is it Actually Worth the HYPE?19:57
61. 61Tiktok made Cursor alternative and Its surprisingly good?16:03
62. 62Llama 4, DeepSeek & the AI Copyright Wars—Dario Weighs In16:43
63. 63DeepSeek’s 500% Profit Margin—How Are They Doing It?6:22
64. 64New OpenAI Agent SDK and More: What Developers Need to Know!16:24
65. 65Deep Research But23:40
66. 66OpenAI’s Responses API: The Easiest Way to Build a RAG System?17:02
67. 67Crash Course: How to Build a Multi-Agent AI System with OpenAI’s SDK28:52
68. 68Forget Chain-of-Thought—This New Tool Might Be Better18:11
69. 69DeepSeek Just Updated—The Best Coding Model Yet?8:58
70. 70Gemini 2.5 Pro Function Calling—All You Need to Know22:24
71. 71GPT-4o’s Upgrade is Kinda Wild (Coding, Images, Unhinged Mode?)19:19
72. 72Q Developer - AI Coder Built for Scale - Code, Review, Ship12:14
73. 73Model Context Protocol (MCP) Explained: The New Standard for AI Tools?15:48
74. 74Forget Next-Word Prediction—LLMs Are Doing This Instead ...25:05
75. 75MCP Security is Still Broken18:02
76. 76This Mixture-of-Agents System Outperforms Manus AI10:34
77. 77Quasar Alpha: This Mystery Coding Model is So Cracked 🔥12:02
78. 78🚀 GPT-4.1 Is Great at Coding, But I Won’t Use It. Here’s Why!18:29
79. 79Google’s AI Studio in 22 Minutes21:28
80. 80OpenAI O3: A Breakthrough in Agentic Coding?25:10
81. 81Learn How to Build Real-World Agents26:40
82. 82Qwen-3 Is Here — The Llama-4 We’ve Been Waiting For!15:39
83. 83Could This Gemini Trick Replace RAG?16:37
84. 84GPT‑4o’s “Yes‑Man” Personality Issue—Here’s How OpenAI Fixed It15:59
85. 85Gemini 2.5 Pro's: The Best Coding Model Just got Better19:31
86. 86Cursor Alternative? Trae AI Just Got Way Better11:40
87. 87Is Deep Agent the Manus/Genspark Killer?6:41
88. 88New Anthropic Study: AIs Hide Plans, Cheat Quietly25:04
89. 89No Chunks, No Embeddings: OpenAI’s Index‑ Long RAG22:38
90. 90OpenAI Codex Agent – Is This the End of Programmers?13:22
91. 91Google's VEO: The Cheapest Way to Use the Best AI Video8:01
92. 92Jules: Google’s Codex Killer?10:33
93. 93From Models to Agentic Applications with Sam Witteveen31:43
94. 94Best Coding Model? I Tested 5 Models.14:59
95. 95New DeepSeek R1 is Really, Really Good Coder12:50
96. 96Gemini 2.5 Pro Beats O3 — Big Drops from ElevenLabs & Qwen11:46
97. 97AI Agents & The Future of Coding: A Conversation with a Googler22:27
98. 98Context Engineering for Building Better Agents23:37
99. 99LocalGPT 2.0: Turbo-Charging Private RAG17:13
100. 100Rogue Agents — When AI Starts Blackmailing — New Study from Anthropic12:53
101. 101Warp: The CLI Agent That Could Replace Claude Code11:58
102. 102Gemini CLI — Google’s Open-Source Coding Agent12:11
103. 103I Gave Devin A Real World Coding Task, Here’s How it Cooked!12:50
104. 104The Only Embedding Model You Need for RAG13:52
105. 105Context Engineering is All You NEED!15:20
106. 106Secret Context Engineering Trick For RAG9:37
107. 107Grok 4—Possibly the Most Powerful Model in the World?11:11
108. 108Kimi K2 - The DeepSeek Moment for Agentic Coding?13:15
109. 109localGPT 2.0 - Building the Best Private RAG System23:04
110. 110Kimi K2 — The Deep Researcher Agent13:18
111. 111ChatGPT Agent Is Here: Your All‑In‑One AI Worker?8:03
112. 112Developers’ Favorite AI Tools in 20259:19
113. 113NEW Qwen 3, Better than Kimi K2?11:32
114. 114NEW Qwen 3 Coder: Did the Benchmark Lie?9:34
115. 115I Built a Voice Agent that Handles my Daily Tasks11:45
116. 116Augment Code: Specs Driven Development For AI Coding Agents9:26
117. 117Master Claude Code Sub‑Agents in 10 Minutes10:04
118. 118GPT-5 - A Good Coding Model?13:52
119. 119Gemini Deep Think: Built for the Hardest Problems14:11
120. 120Horizon: OpenAI’s Secret Open-Weight Model?8:56
121. 121GPT-5 Coding Agent vs. Real App: What It Actually Built in Cursor10:15
122. 122GPT-5: The Most Polarizing Model14:15
123. 123LangExtract + RAG: Smarter Retrieval with Metadata Filtering12:24
124. 124Not all models providers are equal10:45
125. 125GPT-OSS Jailbreak with this Simple Trick11:42
126. 126Finally! A Standard for AI Coding Agents (Agents.md Explained)12:03
127. 127Web Scrapping Made Easy with This MCP12:12
128. 128Claude for Chrome: Agentic Browsing is Here14:24
129. 129Embedding Gemma: On-Device RAG Made Easy11:10
130. 130Super Agent: An Agent That Builds Its Own Tools13:29
131. 131Agent Client Protocol : The “New MCP” for IDEs and Coding Agents8:44
132. 132OpenAI Just Dropped A New Coding Model for Developers9:33
133. 133How Good Is GPT-5 Codex? I Built an App12:04
134. 134Claude Code Downgrade? Here’s What Actually Happened9:54
135. 135Building tools for agents — with agents17:01
136. 136Qwen 3 Omni — The Open AI Model That Does It ALL15:01
137. 137Sonnet 4.5 Is Here—And It’s a Beast at Coding12:40
138. 138OpenAI’s Sora 2 Can Talk—and Follow Physics11:28
139. 139Claude Code 2.0: The Only Guide You Need12:26
140. 140Vibe Planning: The Smarter Way to Code with AI Agents12:45
141. 141OpenAI Just Confused Everyone... Again8:12
142. 142AI Agents Need Context Engineering, Not Bigger Prompts14:15
143. 143Ling 1T Model: Too Good to Be True?9:48
144. 1443 New Context Engineering Skills for Agents14:10
145. 145Can AI Replace Web Developers?9:55
146. 146Vibe for Wordpress - 10Web new AI Builder (no code)7:51
147. 147Haiku 4.5 Is Here—And It’s a Beast at Coding9:23
148. 148The Ultra-Fast and Local Speech Dictation System6:14
149. 149OpenAI Joins the Browser War with ChatGPT Atlas12:37
150. 150Claude Desktop Is Now Even Better — Talk to Claude, Code in Browser, and More!11:44
151. 151Build Custom Skills for Claude Agents14:28
152. 152This Open AI model is a Coding Beast21:51
153. 153What Makes Qwen 3 Max Thinking So Weird?10:35
154. 154Llama.cpp Gets a New Web UI10:48
155. 155Google Goes All-In on Vibe Coding with AI Studio13:55
156. 156No More IDEs - Is this the End?10:30
157. 157Did Google Just Kill RAG? Gemini’s New File Search13:43
158. 158Is This the End of MCP for AI Agents?12:09
159. 159Google File Search API - Managed RAG for Everyone13:37
160. 160Gemini 3 Pro: It was worth the wait!!!21:53
161. 161Google Antigravity - Did Google Just Killed Cursor?13:32
162. 162""I have never felt this much behind as a programmer" - Andrej Karpathy6:06
163. 163Claude’s Best Hidden Feature9:28
164. 164Ralph Wiggum - Ship Code While you Sleep!9:01
165. 165The Holy Grail of Intelligence - Explained.14:48
166. 166Claude Cowork: First General Agents from Anthropic12:14
167. 167Stop Using AI Agents Like Chatbots. (Do This Instead)10:17
168. 168"Don’t Outsource Your Thinking" to your Agent11:15
169. 169OpenAI Codex App: Claude Cowork Killer?9:23
170. 170OpenAI is Bleeding...10:03
171. 171Claude Code's New Agents Team Are Absolutely Insane12:04
172. 172The Easy Way to Get Started with OpenClaw (No Setup Required)8:02
173. 173Minimax-Agent: The Ultimate Open-Source "Workhorse" Model9:06
174. 174How OpenClaw Works: The Real "Magic"7:29
175. 175The "Last Mile" Problem of AI Coding (And how to solve it)7:59
176. 176Codex-Spark: OpenAI Just Broke the Speed Limit (1,000 Tokens/s)9:41
177. 177The 100x AI Breakthrough No One is Talking About15:16
178. 178Why OpenAI Just "Acquired" The Biggest Open Source Agent10:12
179. 179Exploration is All You Need!13:10
180. 180Anthropic Just Killed Tool Calling13:40
181. 181Gemini 3.1 Pro: The model no one expected9:20
182. 182Open Source Backend for AI Agents11:04
183. 183The AI Model Doesn't Matter Anymore17:22
184. 184This 30-Year-Old Pattern Fixes AI Agents16:47
185. 185A Sad Day for Open Source AI11:20
186. 186GPT-5.4: Everything You Need to Know11:20
187. 187The “I Could Build That” Illusion10:09
188. 188Claude Code 2.0: Massive Upgrade with Agent Loops16:14
189. 189Opus just got caught ...12:15
190. 190Claude is taking over...7:36
191. 191Anthropic Just Solved Long Context7:02
192. 192Anthropic Made Their OpenClaw8:23
193. 193AIStudio: Major upgrade with Auth and Database8:28
194. 194Gemma 4 Vision Agent | Object Detection + VLM Pipeline12:42
195. 195Codex Just Became the Everything App9:45
196. 196Claude Design Is Here — Figma Stock Crashed10:56
197. 197RAG is Dead. Again. (Claude Agent SDK + Memory)12:33
198. 198Save 98% on AI Agent Tokens With This One Trick12:31
199. 199What is an Agent Harness? and How to build a great one!20:33
200. 200DeepSeek Just Killed Visual Reasoning (And It's 10× Cheaper)15:05
201. 201Google's New Agent Platform Explained29:54
202. 202Rethinking Agents - Harness is All you Need?14:14
203. 203Agent CLI for Scrapping for Anything14:25
204. 204Thinking Machines Just Solved Real-Time AI Interactions!14:21
205. 205The Model Doesn't Matter. The Harness Does.17:35
206. 206Minimax Mavis Agent: The Verifier Pattern Changes Everything15:37
207. 207Google I/O 2026 Analysis - What You Missed37:32
208. 208Intelligence is Getting MORE Expensive (Google I/O 2026, with Sam Witteveen)38:09
209. 209ElevenLabs 's No-Code Voice Agent Builder12:55
210. 210Stop Building Agent Loops. This Is Replacing Them24:50
211. 211Prompt Caching: Cut Your AI Cost by 90%15:29
212. 212Claude Opus 4.8: Here is Everything that Changed15:11
213. 213Anthropic's Dynamic Workflows: What Everyone Gets Wrong!17:24
214. 214Claude Can Now Build Its Own Harness... For Every Task13:25
215. 215Claude is Building Itself...15:50
216. 216Loop Engineering: Why Everyone is Talking About Agentic Loops?11:04
217. 217Is Claude Fable 5 Worth the Hype? (Technical Deep Dive)14:47
218. 218Does LLM Council/Fusion Actually Work?10:21
219. 219This 284B Model Shouldn't Fit On Your Laptop. It Does20:59
220. 220GPT 5.6 Mythos Level Intelligence11:51
221. 221Design Agents That Run for Days!14:51
222. 222This Tiny Engine Makes Huge AI Models Run Locally (Colibri)12:48
223. 223Inkling: Why Thinky's Open Model May Change "Everything"11:22
224. 224Kimi K3 Explained!9:28
225. 225Opus 5 First Impressions: Anthropic Cooked Again!11:15
226. 226Kimi K3: Can It Beat Opus 5?10:05
227. 227AI Distillation Explained: Why It’s So Misunderstood!15:56