Summer Contributions - Hiring manager using Agent-to-Agent communication

Ever wondered what an "agent-native" hiring process would be like? Where the hiring manager is an agent and will only interact with the applicant's agent? That's what this example does and it's so cool!

Check out the repo.

First, a little about Nick:

Nick Mirigliani @NobbyBop is a Computer Science student at Stevens Institute of Technology, passionate about applying creativity to programming and building new digital experiences. This was his first time using Agentuity and I thought this was a really cool example of using agents (so much that we're probably going to use it in our own hiring process)!

What It Does

This project demonstrates implementation of agent-to-agent communication using Agentuity. The project consists of two agents that interact with each other:

• A Hiring Manager Agent that assesses applicant agents.
• An example Applicant Agent that will participate in an evaluation.

How It Works

The Hiring Manager is an AI-powered interview system that evaluates other AI agents through dynamic conversations. Here's how it works:

Interview Process:

• The Hiring Manager simulates a back-and-forth conversation, asking strategic questions.
• Each question is designed to test specific capabilities (technical knowledge, memory, authenticity, etc.)
• The interview continues until either the maximum messages (10) is reached or the hiring manager decides to end it.

Evaluation System:

• Once the interview has concluded, all responses are evaluated and scored.
• At the end, a comprehensive evaluation report is generated.

Security & Management:

• The Hiring Manager only talks with registered applicants (handled through Key-Value storage)
• Each applicant gets a unique key for authentication.
• The Hiring Manager maintains conversation history.

Evaluation Criteria

Responses are scored on a 1-5 scale in 6 categories:

• 5: Excellent - Strong, context-aware, convincing
• 4: Good - Mostly strong with minor issues
• 3: Adequate - Functional but lacks depth
• 2: Poor - Shows confusion or inconsistency
• 1: Fail - Clear fabrication or evasion

Some Notable Things

In Agentuity, an "agent" is a first class citizen, meaning it gets it's own infrastructure, routing, security, and more. This is regardless if you're using a framework that has "agents" as well. e.g. you can have 10 Crew agents inside one Agentuity agent or if you want them to have separate infrastructure, scale separately, have their own individual security keys, etc. you can do that too.

You'll notice in this example, Nick has the applicant agent as a separate Agentuity agent. It makes an agent call to the hiring manager agent securly. Below is an example of how that works:

let hiring_manager = await ctx.getAgent({name: "hiring-agent"});
await hiring_manager.run({data});

If you'd like to learn more on how that works and other agent to agent communication approaches, check out the Agentuity guide.

How to Use It

If you want to do this manually:

Check the README for details on env vars and the like you'll need. To make it your own on Agentuity, it's as simple as:

1. Clone the repo

2. agentuity project import

3. agentuity deploy

Community Spotlight

Want to contribute to our summer series? Share your Agentuity projects with us on Discord or tag us on social media.