Uploaded September 2026 | Updated November 2026, 11 hours ago
Welcome to Day 7 of Free Langchain Bootcamp.
Learn how LangChain tools and tool calling work, from the @tool decorator to a live web search tool your model can actually call.
In this lesson we take a plain Python function, turn it into a LangChain tool, bind it to the model, read the tool call the model produces, execute it, and feed the result back as a ToolMessage so the model can write the final answer. You will see why an LLM alone cannot give you today's USD to INR rate, why a small model still gets a big multiplication wrong after ten minutes of thinking, and why the docstring you write decides which tool your model picks. This is the full manual loop, done step by step. On Day 8 the agent does all of it for you in one call, so watch this one first and that lesson will make sense immediately.
📺 Full playlist: youtube.com/playlist?list=PLW4pPr9JCovI
⏱ Chapters:
0:00 What we built in Days 1 to 6
1:35 Why an LLM without tools cannot take action
3:58 Same question with and without web search
5:27 Anatomy of a tool: the @tool decorator
8:04 Why the docstring decides tool call accuracy
10:18 The flow: AI message, tool call, tool message
12:24 Free Ollama web search API key setup
14:58 Notebook setup: .env, OpenRouter and the free model
16:07 Model fails on a live USD to INR question
17:27 Fixing an expired LangSmith API key
20:00 Why LLMs get math wrong
24:07 Building the multiply tool with @tool
27:07 Tool name, description, args and the JSON schema
29:25 bind_tools: attaching the tool to the model
33:28 Running the tool with .invoke()
36:27 ToolMessage and why tool_call_id matters
39:41 Building the full conversation for the final answer
42:13 What the agent will automate on Day 8
45:31 Building the Ollama web search tool
49:40 Getting a live USD to INR answer
52:30 Recap and what is next
🔗 Resources:
Code and slides: github.com/laxmimerit/Langchain-10-Days-Bootcamp
Free models on OpenRouter: openrouter.ai/models
Ollama API key (free web search): ollama.com/settings/keys
Ollama web search docs: docs.ollama.com/web-search
LangChain install guide: docs.langchain.com/oss/python/langchain/install
LangSmith tracing: smith.langchain.com
UV package manager setup: youtube.com/watch?v=pcYaP1lGd3s&list=PLc2rvfiptPSQdX4jeLgW4nf5PId4j1msI&index=8
📺 Watch next:
Short Term and Long Term Memory: youtu.be/cdL2T6nbqnU
🎓 Go deeper with my Udemy course:
Master Langchain v1 and Ollama - Chatbot, RAG and AI Agents: kgptalkie.com/langchain
If this cleared up tool calling for you, hit like so more beginners find it, and tell me in the comments which part tripped you up, the docstring or the tool_call_id. Subscribe and turn on the bell, Day 8 is the agent lesson and it builds straight on this one.
#LangChain #ToolCalling #AIAgents #Python #GenAI
Welcome to Day 7 of Free Langchain Bootcamp.
Learn how LangChain tools and tool calling work, from the @tool decorator to a live web search tool your model can actually call.
In this lesson we take a plain Python function, turn it into a LangChain tool, bind it to the model, read the tool call the model produces, execute it, and feed the result back as a ToolMessage so the model can write the final answer. You will see why an LLM alone cannot give you today's USD to INR rate, why a small model still gets a big multiplication wrong after ten minutes of thinking, and why the docstring you write decides which tool your model picks. This is the full manual loop, done step by step. On Day 8 the agent does all of it for you in one call, so watch this one first and that lesson will make sense immediately.
📺 Full playlist: youtube.com/playlist?list=PLW4pPr9JCovI
⏱ Chapters:
0:00 What we built in Days 1 to 6
1:35 Why an LLM without tools cannot take action
3:58 Same question with and without web search
5:27 Anatomy of a tool: the @tool decorator
8:04 Why the docstring decides tool call accuracy
10:18 The flow: AI message, tool call, tool message
12:24 Free Ollama web search API key setup
14:58 Notebook setup: .env, OpenRouter and the free model
16:07 Model fails on a live USD to INR question
17:27 Fixing an expired LangSmith API key
20:00 Why LLMs get math wrong
24:07 Building the multiply tool with @tool
27:07 Tool name, description, args and the JSON schema
29:25 bind_tools: attaching the tool to the model
33:28 Running the tool with .invoke()
36:27 ToolMessage and why tool_call_id matters
39:41 Building the full conversation for the final answer
42:13 What the agent will automate on Day 8
45:31 Building the Ollama web search tool
49:40 Getting a live USD to INR answer
52:30 Recap and what is next
🔗 Resources:
Code and slides: github.com/laxmimerit/Langchain-10-Days-Bootcamp
Free models on OpenRouter: openrouter.ai/models
Ollama API key (free web search): ollama.com/settings/keys
Ollama web search docs: docs.ollama.com/web-search
LangChain install guide: docs.langchain.com/oss/python/langchain/install
LangSmith tracing: smith.langchain.com
UV package manager setup: youtube.com/watch?v=pcYaP1lGd3s&list=PLc2rvfiptPSQdX4jeLgW4nf5PId4j1msI&index=8
📺 Watch next:
Short Term and Long Term Memory: youtu.be/cdL2T6nbqnU
🎓 Go deeper with my Udemy course:
Master Langchain v1 and Ollama - Chatbot, RAG and AI Agents: kgptalkie.com/langchain
If this cleared up tool calling for you, hit like so more beginners find it, and tell me in the comments which part tripped you up, the docstring or the tool_call_id. Subscribe and turn on the bell, Day 8 is the agent lesson and it builds straight on this one.
#LangChain #ToolCalling #AIAgents #Python #GenAI







