<aside> 🦥

Sources

Building Effective AI Agents – Anthropic

https://huggingface.co/learn/agents-course/en/unit1/agent-steps-and-structure

</aside>

<aside> 🦥

How "agentic" something is comes down to how independent it is. There are 3 main levels 👇

</aside>

1. Non-agentic LLMs (The Basics)

<aside> 🦥

These are basically your normal LLMs/AIs. ChatGPT, Gemini, Claude, Llama, etc.

</aside>

<aside> 🦥

• What they are: AI systems that can only respond with text based on their training data.
• How they work: They analyze your question and generate a response using patterns from their training
  • Obviously it’s more detailed, but this isn’t about LLMs…
• Real-world example: Using basic ChatGPT to answer a question
• Limitations:
  • Can't look up new information online
  • Don't remember conversations after you close the chat
  • Can't use tools or take actions in the real world
  • Like talking to someone who's knowledgeable but cut off from the outside world </aside>

2. AI Workflows (The Middle Ground)

<aside> 🦥

These are like LLMs with special abilities. They have access to specific tools and information.

</aside>

<aside> 🦥

• What they are: LLMs connected to specific tools and data sources
• How they work: They can use predetermined tools when instructed, but follow fixed patterns
• Real-world example:
  • ChatGPT with browsing capability looking up today's weather
  • Claude accessing your Google Drive to find a specific document when you ask
• Key features:
  • Can access external information when specifically asked.
  • May remember previous conversations
  • Use tools, but usually need direct instructions for which tool to use: “Search the internet to find blah blah blah”
  • AI workflows are like a helpful assistant who can look things up and use basic tools, but needs step-by-step guidance </aside>

3. AI Agents (The Advanced Systems)

<aside> 🦥

These are like digital assistants with initiative. They can decide what to do and how to do it.

</aside>

<aside> 🦥

• What they are: Systems that can observe, plan, decide, and act toward goals with minimal supervision
• How they work: They break down complex tasks, decide which tools to use, and learn from results
• Real-world example:
  • An AI system that, when asked to "research vacation options for my family," automatically searches travel sites, checks your calendar for available dates, compares prices, and presents options.
  • Claude Code, which can plan and execute complex coding tasks from a simple request
• Key features:
  • Can create and follow multi-step plans without guidance at each step
  • Chooses appropriate tools autonomously
  • Adapts when the initial approach doesn't work
  • Maintains memory of what it has learned and accomplished
  • Like having a proactive assistant who understands your goals and figures out how to achieve them on their own </aside>

AI Workflows Vs AI Agents

<aside> 🦥

Decision-making Process

• Workflows: Follow predefined paths with limited decision points
• Agents: Make autonomous decisions about what actions to take and when </aside>

<aside> 🦥

Goal Understanding

• Workflows: Execute specific commands as given
• Agents: Understand higher-level goals and determine how to achieve them </aside>

<aside> 🦥

Handling Complexity

• Workflows: Excel at routine tasks with clear steps
• Agents: Can tackle open-ended problems requiring creative solutions </aside>

<aside> 🦥

Adaptation

• Workflows: May struggle when encountering unexpected situations
• Agents: Can adapt their approach when the initial plan fails </aside>

<aside> 🦥

User Interaction

• Workflows: Require explicit instructions at each step
• Agents: Need only high-level directions and can fill in the details </aside>