flintspark Book a demo

Learn about AI agents

Learn is flintspark's plain-language guide to AI agent concepts. It explains the ideas behind teaching an agent by showing it a task once, keeping people in the loop and recording what the agent did, and says where flintspark, whose agent is Flint, fits. Start with the glossary.

Demo: the supplier's confirmation page after Flint, flintspark's agent, filed a diesel request, with a written reason for each region
A frame from the energy demo film: the request Flint filed, with a written reason for each region. The terms on these pages describe what happened in that run. The supplier is fictional.

Where should I start?

With the AI agent glossary. It defines eight terms in one or two sentences each, and says how flintspark uses each one and where on this site to read more.

The terms are teach by demonstration, skill, verified run, execution route, approval grant, human in the loop, audit trail and redaction.

A sketch of an entry, not a screen from the product.

What do these ideas look like?

Three of the glossary's terms, as they appear in a run. Each sketch uses the energy demo's names and links to its definition.

Teach by demonstration and redaction
Approval grant and human in the loop
Audit trail: each entry carries the hash of the one before it

What is coming next?

Four longer guides are being written. They are not published yet, so they are named here without links.

Last updated

  1. Teach by demonstrationBuilding an agent's skill by doing a task once while it is recorded
  2. Human in the loopHow people answer questions and approve key actions inside an automated workflow
  3. Audit trail for AI agentsWhat to record, what tamper-evident means and what it does not prove
  4. Computer-use agentsHow an AI that operates software through its screen works, and where it fails

See Flint run

Bring one workflow you'd rather not do by hand. We'll show Flint learning one like it, and stopping before anything it can't undo.

We reply from hello@flintspark.ai.