Train an AI agent to work the way you do

Every agent starts out knowing its model, not your business. Training a Crewdle Connect agent is how you close that gap, and it works the way onboarding a new hire does: give it a real task, look at what comes back, correct it in plain words, and tell it to remember. No fine-tuning, no code, no settings screens.

This guide trains one real agent from scratch: a proposal writer for a small video production studio. By the end, a one-line request produces a client-ready Word document in the house format, with the house pricing math applied on its own. Your part takes about 15 minutes; the agent does the rest.

Step 1: Hire the agent and be honest about what it does not know

In Connect, open Agents, click Hire an agent, and describe the job in your own words. Say what you plan to teach: we wrote "I will teach you our house format, tone, and pricing rules as we go" right in the description. The wizard proposes a ready-made setup, then asks a few refining questions. Answer the ones you have answers for and skip the rest; we filled in the tone question and deliberately left format and pricing blank, because those are exactly what the training will cover.

Put the teaching plan in the job description. What setup does not cover, training will.

How to hire a trainable agent

1. In Agents, click Hire an agent, name it, and describe the job, including what you will teach it as you go.
2. Click Suggest agents and pick the proposed setup, or describe changes and let the options revise.
3. Answer the refining questions you can, and skip the ones you would rather teach by example.

Step 2: Hand over examples, if you have them

The last wizard step asks for documents, and it suggests exactly the right kind for the job it now understands: recent winning proposals, a rate card, a template. Anything you drop here becomes reference material from day one, which shortens training. We added nothing, on purpose, to show that conversation alone gets you there. If you have good examples, upload them; the two approaches stack.

The wizard suggests the reference documents worth uploading. Examples shorten training, but they are optional.

Step 3: Ask for a first draft, and read it like an audition

Give the agent its first real task. We asked for a proposal for a dental clinic: a two-day shoot plus editing, delivered as a Word document. Minutes later the draft was in the conversation, and it did something telling: it left the Investment section as a placeholder instead of inventing prices it had never been given. That is the point of a first draft. Its gaps are your syllabus; they show you exactly what to teach.

The first draft arrives as a Word document, with an honest gap where the pricing rules should be.

Step 4: Correct it, then say "save these as your standing rules"

This is the training move. Reply with corrections concrete enough to act on, and end with one instruction: save these as your standing rules for every future proposal. Ours was a single message: five sections in a fixed order, shoot days at $3,500 per day, editing at 40% of the shoot fee shown as line items, one page of plain language, and a closing with a 50% deposit and 30-day validity. Then we asked it to update the draft to match.

The agent confirmed the rules were saved and reworked the document: 2 shoot days at $3,500 is $7,000, editing at 40% is $2,800, total $9,800. It even hit a small verification error mid-update and fixed it on its own, which is what an autonomous agent does with a snag.

One message turns feedback into standing rules. The agent saves them and reworks the current draft to match.

How to train with feedback

1. Point at what is wrong and say what right looks like: sections, numbers, tone, length.
2. End with "save these as your standing rules for every future proposal" (or report, or reply), so the correction outlives the conversation.
3. Ask it to redo the current deliverable, and check that the fix landed before moving on.

Step 5: Open its memory and read what it learned

Training you can read is training you can trust. Open More options in the agent's header and choose View memories: every lesson the agent keeps lives there as a small text file you can open. Ours had written proposal-role-best-practices.md, and inside were the five sections, the day rate, the editing percentage, and the closing terms, exactly as taught, alongside working notes it added on its own.

This panel is your control surface for the agent's behavior. If a memory is wrong, outdated, or something you never meant to teach, say so in the conversation and the agent revises it; check the file again after.

Open More options, then View memories, and read the rules for yourself. This file is why the training sticks.

How to review an agent's memory

1. Open the agent, click More options in its header, and choose View memories.
2. Open a memory file to read exactly what the agent has saved.
3. Wrong or missing rule? Correct it in the conversation, then check the file again.

Step 6: Test it on a fresh task, repeating nothing

Training you have not tested is a hope. Send a new request and leave every rule out of it. Ours was one line: a realty office, a three-day shoot across their listings, plus editing. The agent priced it entirely from its saved rules: 3 shoot days at $3,500 is $10,500, editing at 40% is $4,200, total $14,700. Nothing in the request mentioned money at all.

The test that matters: a one-line request, and the trained rules applied without being repeated.

Step 7: Open the finished document

Click Open on the file card, or find everything later under Files in the agent's header, in the Generated tab. Our second proposal came back with the five sections in order, the Investment table with the right math, the deposit and validity terms at the close, and exactly one page. That is the trained format, applied unprompted. From here, every proposal costs one sentence.

The document proves the training: house sections, house math, one page, no reminders.

Keep training it

Training is not a phase; it is how you manage the agent from now on. When your rates change, tell it to update its standing rules. When a draft misses, correct it once and make the correction a rule. Each pass compounds, so the agent is worth more every week you work with it.

Not every lesson arrives as a tidy instruction, though. After a working session full of small corrections, click Memorize this conversation in the agent's header: the agent distills the exchange into memory on its own, then posts a note of what it kept. Ours banked the format and pricing rules, the tone preference, and project notes for both clients, which appeared as a new file under View memories.

One click banks the whole conversation: the agent saves the durable lessons and tells you exactly what it kept.

What it learns also travels. A trained agent's memory files, procedures, and knowledge ship as the knowledge pack when you rent it out to other organizations, so review its memories before you list it and you know exactly what you are shipping. Its results can be proven on a public leaderboard in the Arena, and the usual oversight applies while it learns: its logs show every action and credit, and a monthly budget caps what it can spend. See Credits and billing.