Infra & cost
Webhook
An HTTP callback an AI service makes to your endpoint when a long-running event completes — async results, agent updates, batch jobs.
01 ——
In plain English
A webhook is a URL the AI service POSTs to when something interesting happens — usually because the work is too long to wait for in a single HTTP request. You hand the provider your endpoint at job-start; they call you back when the job is done.
Why AI APIs need them:
- Long-running generation — video, image batches, fine-tuning jobs, deep-research agents
- Streaming alternative — you can stream responses, but webhooks are better for "fire and forget" use cases
- Batch processing — Anthropic Message Batches, OpenAI Batch API return via webhook or polling
- Agent events — long-running agent runs push progress updates as webhook events
Common patterns:
- Single completion callback — one POST when the job is done
- Streaming events — multiple POSTs as the agent makes progress
- Signed payloads — providers sign requests so you can verify they're real
Securing them: Validate the signature header (every major provider supports it). Otherwise anyone who guesses your URL can POST fake events.
02 ——
Related terms
API
An interface that lets developers send requests to an AI model and get responses programmatically — the way most AI tools talk to LLMs.
Streaming
Sending an AI model's response token-by-token as it's generated, so the user sees text appear immediately instead of waiting for the full reply.
Latency
The time it takes an AI model to respond to a request — from when you hit send to when the first or final word appears.
Rate Limit
A cap on how many requests or tokens a user can send to an AI API in a given window — used to manage cost, capacity, and abuse.
Inference
The process of running a trained AI model to generate a response — as opposed to training the model.
Agentic Workflow
A multi-step task where an AI agent autonomously decides the steps, uses tools as needed, and works toward a goal with minimal human steering.
Back to glossaryLast reviewed May 2026