50 min
Lesson 4 of 6
Assistants API & Tools
From the course OpenAI API Masterclass
The Assistants API
The Assistants API provides stateful, multi-turn conversations with built-in tools for code execution, file search, and function calling.
Creating an Assistant
const assistant = await openai.beta.assistants.create({
name: "Data Analyst",
instructions: "You analyze data and create visualizations.",
model: "gpt-4o",
tools: [
{ type: "code_interpreter" },
{ type: "file_search" }
],
});
Threads & Messages
Conversations happen in threads that persist across sessions:
const thread = await openai.beta.threads.create();
await openai.beta.threads.messages.create(thread.id, {
role: "user",
content: "Analyze the uploaded sales data",
});
const run = await openai.beta.threads.runs.create(thread.id, {
assistant_id: assistant.id,
});
Built-in Tools
- Code Interpreter: Executes Python code in a sandbox. Processes files, generates charts, performs calculations
- File Search: Indexes uploaded files with vector search for retrieval
- Function Calling: Define custom functions the assistant can invoke
Function Calling Pattern
const assistant = await openai.beta.assistants.create({
model: "gpt-4o",
tools: [{
type: "function",
function: {
name: "get_weather",
description: "Get current weather for a city",
parameters: {
type: "object",
properties: {
city: { type: "string" },
unit: { type: "string", enum: ["celsius", "fahrenheit"] }
},
required: ["city"]
}
}
}]
});
Run Lifecycle
Runs go through states: queued → in_progress → requires_action → completed. Poll or use streaming to track progress and handle tool calls.
Key Takeaways
- •Assistants maintain stateful conversations across sessions via threads
- •Code Interpreter executes Python for data analysis and visualization
- •File Search enables RAG over uploaded documents automatically
- •Function calling connects assistants to external APIs and data
0 of 6 lessons marked complete
Progress is kept in this browser only.