explaingit

jack-and-jill-ai/jack_and_jill_ai_guides

17Audience · pm founderComplexity · 1/5Setup · easy

TLDR

A public collection of internal AI agent design frameworks from a recruiting startup with 200k users, sharing practical guides for defining agent identity, memory structure, and workflow file layout based on real production experience.

Mindmap

mindmap
  root((ai guides))
    What it is
      Internal frameworks
      Open sourced guides
    Topics covered
      Agent persona design
      Memory structure
      Workflow setup
    Company context
      Recruiting marketplace
      Fleet of named agents
      200k users
    Audience
      AI builders
      Startup founders
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Design a persistent AI agent persona with a defined identity, structured memory, and workflow using the published framework as a template.

USE CASE 2

Adapt the agent file layout the company uses across their production fleet to organize your own team's AI agents consistently.

Getting it running

Difficulty · easy Time to first run · 30min

In plain English

Jack and Jill AI Guides is a public collection of internal frameworks and documents from Jack and Jill, a company building an AI-powered recruiting marketplace. They run a fleet of AI agents internally to power both their product and their operations, and this repository is where they publish what they learn from that work, open sourced for anyone else building with AI agents. The first guide in the repository is a framework for designing and building persistent AI agent personas from scratch. It covers defining an agent's identity, structuring its memory, setting up its workflows, and the file layout the company uses across their own agent fleet. The document is positioned as the kind of internal reference that took real trial and error to develop, shared so others do not have to repeat the same process. The company itself runs two public-facing agents: Jack is a free career agent that helps job seekers, and Jill is a recruiter-facing agent for employers. Behind those two, the team also runs additional internal agents named Juno, Joe, James, Jules, Jessica, and Janice, among others. These internal agents handle product and operational tasks that are not directly user-facing. The README notes the company has more than 200,000 users and ships over 60 pull requests per week. The repository is sparse at the moment, with one guide published so far and more planned as the team continues building. It functions as a public log of practical lessons from a team building AI-native software at a fairly high pace. If you are trying to understand how to structure an AI agent system for a company's internal functions, this repository offers one team's concrete approach. The guides are written in plain text and do not require any specific programming language or framework to read and apply.

Copy-paste prompts

Prompt 1
Using the Jack and Jill AI agent persona framework, help me design a customer support AI agent with a distinct identity, structured memory, and a workflow for handling refund requests.
Prompt 2
Walk me through the agent file layout from jack_and_jill_ai_guides and adapt it for a sales outreach AI agent that needs to remember past conversations with each prospect.
Prompt 3
Based on the Jack and Jill AI agent design guide, what are the key decisions I need to make before building my first persistent AI agent and what pitfalls should I avoid?
Prompt 4
Use the Jack and Jill agent persona framework to help me define the identity, memory structure, and workflow for an internal operations agent that handles scheduling and task routing.
Open on GitHub → Explain another repo

← jack-and-jill-ai on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.