---
title: "A talent agent built an AI chatbot with Make to answer client questions and grew from about 100 clients to 300 without new staff"
person: "U-Shin Kim (U-Shin Group Artists)"
tools:
  - "Make"
  - "ChatGPT"
country: "Austria"
category: "Make money"
category_slug: make-money
proof:
  - label: "Read Make's customer story"
    url: "https://www.make.com/success-stories/talent-agent-challenged-the-hollywood-status-quo-with-make"
date: 2026-11-04
date_posted: 2026-02-13
as_claimed: "The result is shown as claimed in the public source linked here. It has not been checked and is not a typical outcome."
url: "https://agentstories.org/stories/u-shin-kim-make-ai-agent-talent-agency.html"
markdown_url: "https://agentstories.org/stories/u-shin-kim-make-ai-agent-talent-agency.md"
site: "Agent Stories by Solopreneur Code"
---

# A talent agent built an AI chatbot with Make to answer client questions and grew from about 100 clients to 300 without new staff

U-Shin Kim (U-Shin Group Artists) · Austria · Make, ChatGPT

## What the agent did

Kim mapped his whole talent representation process and built Ari, an AI agent chatbot made with Make, which answers routine client questions, down to outfit suggestions for premieres. He also uses Make with ChatGPT for outbound emails and an actor application workflow.

## Result, as claimed

Make says his client count went from about 100 to 300 with no new staff, and that he saves 15 hours a week. He is aiming for up to 1,000 clients, which is a goal, not a result.

> The result is shown as claimed in the public source linked here. It has not been checked and is not a typical outcome.

## Proof

- [Read Make's customer story](https://www.make.com/success-stories/talent-agent-challenged-the-hollywood-status-quo-with-make)

## Try it yourself

Build the answer bank and hand-over rules for a chatbot that answers your clients' routine questions.

Prompt:

```text
I run a [type of service business] with about [number] clients. Here are the questions clients ask me most often, with my usual answers: [paste 15 to 30 of them]. Turn this into (1) an answer bank written in my tone, (2) a list of questions the chatbot must never answer and must pass to me, such as fees, contracts and complaints, (3) a short hand-over message for those cases, and (4) ten test questions, including two tricky ones, to check it before clients use it.
```

Tools:

- [Make](https://www.make.com/en/register?pc=anferneeck): Named in the story. He built the Ari chatbot and his workflows with it
- [ChatGPT](https://chatgpt.com): Named in the story for his outbound emails

Steps:

1. Collect a month of the questions clients actually send you.
2. Run the prompt and correct every answer yourself.
3. Load the answer bank into the chatbot you build and connect it to your message channel.
4. Read the chat logs every week and add the questions it got wrong.

Watch out: Clients may treat a chatbot answer as your promise, so keep fees, contracts and anything personal with a person.

## More

- [This story as a web page](https://agentstories.org/stories/u-shin-kim-make-ai-agent-talent-agency.html)
- [More stories: Make money](https://agentstories.org/categories/make-money.html)
- [More stories from Austria](https://agentstories.org/countries/austria.html)
- [More stories that use Make](https://agentstories.org/tools/make.html)
- [More stories that use ChatGPT](https://agentstories.org/tools/chatgpt.html)
- [Every story, indexed for AI assistants](https://agentstories.org/llms.txt)
