Hiring a freelancer? Here’s how to use AI to vet them

Looking for a freelancer to work on your project? Don’t be fooled by their AI-written bios, cover letters, or proposals.

Use AI to pressure-test their proposals and reveal the truth behind them.

Below are the four tests — and how to automate them.

1. Test for specificity — this reveals vague statements

People who understand a problem naturally speak in specifics. People who don’t rely on safe, vague, padded lines.

Ask your AI tool:

Extract every sentence in the proposal that claims:
- experience
- capability
- familiarity
- past success

Then check for each:
- What's missing?
- What question is avoided?
- What example should be here?

What this reveals:

• Missing scope details → they don’t understand the job
• Missing tool details → they haven’t used the tools
• Missing timeline details → they can’t plan the work
• Missing ownership details → they supported, but didn’t lead

Vague lines usually hide missing skills.

2. Test for context awareness — this reveals template reuse

Most AI-written proposals fall apart when you ask them to match the job.

Ask AI:

Find sentences in the proposal that should reference the job post but do not.

For each:
- What job detail is missing?
- What should they have copied or echoed from the job post?
- What does this say about the freelancer?

This exposes:

• Proposals that ignore your project needs
• Proposals that skip your constraints
• Reused opening lines
• Generic skill claims
• Proposals written before reading your post

Good freelancers respond to your job. Weak ones reuse old pitches.

3. Test for experience grounding — check if their experience actually supports their proposal

Most people check whether the freelancer’s experience “matches” the project.

It’s better to check whether their experience actually supports the promises they make in the proposal.

Ask AI:

Find claims in the proposal that are not supported by:
- past project examples
- tool usage
- deliverables
- measurable outcomes

For each unsupported claim:
- What kind of project would normally support it?

This uncovers people who:

• Sound senior but worked junior roles
• Add irrelevant projects
• Have no portfolio
• Use mismatched tech
• Rely on outdated work

This separates real and current experience from pretend and stale experience.

4. Generate a failure-mode summary — this reveals the true risk of hiring them

Most founders hire by looking for strengths. Experienced founders hire by looking for failure patterns.

Ask AI:

Summarize where this proposal fails:
- specificity
- context
- experience grounding

Which failure is most costly for this project?

Does this freelancer understand the work well enough to interview?

Now you have:

• A clear risk picture

• A clear yes/no

• A clear reason behind it

This is the clarity good systems give you.

You’ve seen the tests. Now here are three simple ways to use them.

5. The easiest version: ChatGPT / Claude

This is the simplest way to run the system.

Here’s how:

1. Paste the job
2. Paste the proposal
3. Paste the prompts

The AI gives you the breakdown right away.

Use this when:

• you want speed
• you don’t need organization
• you’re doing one-off reviews

6. A simple version inside Notion

If you already keep your jobs and proposals in Notion, this makes the process smoother.

Do this:

1. Make a Notion database
2. Add a page for each proposal
3. Paste the job and the proposal
4. Select everything → Ask AI
5. Paste the prompts

Use Notion when:

• you already work in Notion
• you want the job, proposal, and review in one place
• you want something simple but more organized than ChatGPT or Claude

7. The automated version: Airtable

Airtable requires more setup, but it automates everything.

You set it up once, and then every new proposal is reviewed automatically:

Here’s how.

Step 1: Create your Airtable base

1. Go to Airtable → Create a new base
2. Name it Freelancer Proposals
3. Add these fields (columns):
   • Name
   • Job Post
   • Proposal
   • AI Review
   • Risk Level
   • Recommendation

Step 2: Add the automation

1. Go to the Automations tab
2. Click Create an automation
3. Choose the trigger:
   • When a record is created
   • Table: Freelancer Proposals

Step 3: AI analysis

1. Click Add action
2. Choose “Generate text with AI” (This opens a text box where you write what the AI should do.)
3. Paste:

JOB:{{Job Post}}
PROPOSAL:{{Proposal}}
\[Insert all 4 prompts from our previous sections here: Prompt 1, Prompt 2, Prompt 3, Prompt 4\]

Step 4: Save the output

1. Add another action: Update Record

2. Choose the same table: Freelancer Proposals

3. Set the record ID to the trigger record

4. In the fields:

   • Map AI Review → from the AI output
   • Map Risk Level → (AI can output “High / Medium / Low”)
   • Map Recommendation → from AI output

Click Save.

Done.

Use Airtable when:

• you get many proposals
• you want a system that runs by itself
• your team needs to see everything in one place

If you only review a few proposals, Airtable isn’t worth the setup.

Use ChatGPT or Notion instead.