β What if you moved your workflows from ChatGPT's website to superagents like Cursor?
- Instead of copying & pasting between browser tabs, your AI understands your entire business context
- No more losing conversation history or starting from scratch every session
- AI that knows your company's strategies, processes, and goals
β What if your entire company used tools like Cursor and shared context, prompts, scripts, and workflows across the whole organization?
- Your marketing team's AI expertise becomes available to your strategy team
- Your operations workflows become templates for the entire company
- Every team member benefits from the collective AI intelligence of your organization
Most companies today: Everyone uses ChatGPT individually, recreating prompts, losing context, working in silos.
What this template enables: Your entire organization becomes an AI-powered organism where knowledge, workflows, and expertise compound across all departments.
π¨βπ» Created by: Seva Ustinov | π± Telegram | π¦ X/Twitter | πΌ LinkedIn
Based on the real-world Elly Analytics workspace implementation
This is the complete framework that enables companies to migrate their entire workforce β developers, marketers, strategists, operations teams, and executives β from isolated AI usage to organizational AI intelligence.
π€ AI-Powered Productivity for Everyone:
- Personal task automation - AI handles routine work while teams focus on strategy
- Custom agent context - Each team member can build specialized AI workflows
- Cross-department intelligence - AI agents understand your full business context
π Unified Knowledge Sharing:
- Department-specific expertise - AI adapts to strategy, product, marketing, operations contexts
- Shared workflows & prompts - Leverage specialized AI tools created by colleagues
- Full organizational transparency - See the complete picture beyond departmental silos
π’ Enterprise-Ready Structure:
- Scalable repository architecture - Each department = separate repo with proper access control
- Professional git workflows - Maintains enterprise standards while enabling AI collaboration
- Battle-tested at scale - Proven with 40+ person team across multiple countries
- Growing companies (10-100+ employees) ready to become AI-first
- Organizations with multiple departments needing structured collaboration
- Leadership teams wanting to unlock AI productivity across all functions
- Companies migrating from traditional tools to developer-quality workflows
This template contains real examples from Elly Analytics showing:
- Strategy - Competitive analysis, business planning, executive decision-making
- Product - Roadmaps, specifications, user research, technical planning
- Sales & Marketing - Campaigns, content creation, sales processes, go-to-market
- Operations - Process documentation, metrics, hiring & recruitment workflows
- Finance - Financial models, projections, unit economics
- Projects - Client project management, team allocation, portfolio tracking
- Department-specific AI behavior - AI adapts expertise per team
- Cross-functional collaboration - Maintain specialization while enabling transparency
- Sophisticated prompt engineering - Pre-built AI workflows for each department
- Repository architecture - Multi-repo structure with proper access control
- Automation scripts - One-command setup and daily sync workflows
- Professional git workflows - Enterprise-ready collaboration patterns
- Strategy documents - Competitive analysis, business model planning
- Product specifications - Technical requirements, user research, roadmaps
- Marketing campaigns - Content creation, lead scoring, campaign tracking
- Operational processes - Hiring workflows, project management, team coordination
π Beyond Basic AI Adoption:
- Most companies use AI for individual tasks
- This framework enables organizational AI intelligence
- AI agents understand your complete business context
πΌ Enterprise-Ready:
- Professional git workflows and access control
- Scalable architecture for growing teams
- Proper separation of concerns between departments
π― Practical Implementation:
- Real examples and templates, not just theory
- Proven workflows refined through daily use
- Clear migration path from traditional tools
No. The key difference is code execution.
Most "AI solutions" are glorified chatbots or knowledge bases that can summarize and search. The real magic happens when AI can write and execute code.
- π Notion/Google Docs = useful for documentation
- ποΈ Knowledge bases = valuable for information storage
- π» AI that writes and executes code = transformative automation
This framework enables your AI to actually do things - process files, run analyses, automate workflows - not just talk about them.
You get a complete, working business intelligence workspace:
ποΈ Ready-to-use structure:
- π 7 department repositories (Strategy, Product, Marketing, Operations, Finance, Legal-HR, Hiring)
- π Dev repositories for technical projects
- π Projects folder for client work
- π Presales materials repository
π€ AI Configuration:
.cursorrulesfiles for each department with specialized AI behavior- Context-switching system ("use marketing context" β AI becomes marketing expert)
- Department-specific prompts and guidelines
π Real examples and templates:
- Hiring pipeline with candidate evaluation frameworks and scoring
- Marketing campaign planning and analysis workflows
- Financial models and investor update templates
- Strategy documents including competitive analysis and business models
- Client project management templates
βοΈ Automation scripts:
- One-click setup for your organization (
setup/clone-all-repos.sh) - Daily sync scripts (
setup/update-all.sh) - Cross-platform support (Windows PowerShell + Mac/Linux Bash)
π― What you can try immediately:
- Ask AI to analyze the example competitor research
- Have AI create a new marketing campaign using existing templates
- Test the hiring evaluation system with sample candidates
- Generate financial projections using the included models
π All based on real company usage - these aren't theoretical templates, they're workflows refined through daily use at a real business.
Think of GitHub as Google Drive, not as a developer tool.
You'll never manually type git commands. Your AI agent handles all the technical stuff. For you, it's just:
- π Folders with your files
- π¬ "Hey, download all updates" β AI syncs everything
- π¬ "Upload my changes" β AI commits and pushes
The benefit: version history, team collaboration, and professional-grade file management that scales.
Gradual migration, department by department:
- Week 1-2: Set up one department (recommend Strategy)
- Week 3-4: Move that department's content over
- Week 5+: Add departments one by one as they see the value
You can run both systems in parallel. Most teams find they naturally stop using the old system because asking your AI agent is faster than searching through web interfaces.
It requires management, like any team member.
From early implementations:
- β 80-90% accuracy on well-defined tasks (with proper setup)
- β Dramatically improves with specific prompts and examples
- β
Gets better over time as you refine your
.cursorrules
Pro tips from early adopters:
- Be very specific in your instructions
- Use examples of good vs bad output
- Include source citations requirements to reduce hallucinations
- For critical tasks, have AI generate code/scripts for verification
Yes, through AI-generated scripts and integrations.
Example workflow:
You: "Connect to our Google Sheets with revenue data and create a monthly summary"
AI: Creates Python script β Connects via API β Updates markdown files β You review output
Enterprise-grade security through GitHub's access controls:
- π’ Organization-level permissions - Control who sees what
- π Repository-level access - Finance team can't see HR data
- π Audit trails - Full history of who changed what
- π Client data isolation - Separate repositories for client projects
AI safety:
- Configure AI to cite sources and avoid hallucinations
- Client-specific data stays in separate repositories
- Use
.cursorrulesto define data handling policies
Absolutely. The framework is designed to be adapted:
For smaller companies (5-20 people):
- Start with 2-3 departments
- Use simpler repository structure
- Focus on core workflows first
For larger companies (100+ people):
- Add more department-specific repositories
- Implement more sophisticated access controls
- Create industry-specific templates
Surprisingly gentle, even for non-technical users:
Week 1: "This is just like having a very smart assistant that can read all our files" Week 2: "Wait, it can also create and edit files automatically?" Week 3: "I'm asking it to automate tasks I never thought were automatable"
Most common "aha moments":
- π― Recruiters: "It can score all candidates consistently using our framework"
- π Analysts: "It can process dozens of call transcripts in minutes"
- π Writers: "It knows our brand voice from reading all our past content"
- πΌ Executives: "I can get cross-department insights without interrupting anyone"
π’ Early Stage, High Interest:
- β Active implementation at Elly Analytics (original company)
- π¬ ~20 companies running early experiments and pilots
- π¬ Strong community interest from founders and business leaders
- π Rapid evolution based on real-world feedback
π οΈ We're building this together:
This template reflects our current best practices, but we're learning fast. Every company that tries this approach discovers new use cases, automation opportunities, and workflow improvements.
π€ Contribution Welcome:
- π Share your .cursorrules - Department-specific AI prompts that work
- π Add integrations - Scripts for connecting your tools (CRM, analytics, etc.)
- π Contribute templates - Industry-specific workflows and examples
- π Report issues - What doesn't work in your environment
- π‘ Suggest improvements - Better folder structures, automation ideas
π§ How to contribute:
- Open issues/PRs on this repository
- Share your customizations and learnings
- Join discussions about best practices
- Help newcomers with setup and troubleshooting
β Pros:
- Get started with proven framework structure
- Benefit from real company experience
- Join active community developing best practices
- Free, open-source, no vendor lock-in
- You'll be an early adopter (exciting but requires some pioneering spirit)
- Some rough edges as we refine workflows
- Best practices still evolving
- You might need to adapt examples to your specific needs
πͺ Perfect if you:
- Want to be at the forefront of AI-first business operations
- Enjoy experimenting and contributing to open source
- Have basic comfort with file management and AI tools
- Want to help shape the future of AI-assisted teamwork
This template includes
.gitignore.templateinstead of.gitignoreso you can see all example folders on GitHub. When setting up your workspace, this will automatically be copied to.gitignoreto properly exclude department repositories.
π€ Use AI to Implement AI-First Workflows
This template is designed for AI-first organizations, so the implementation should be AI-assisted too!
Instead of following the manual steps below:
- π Create a new folder for your company workspace
- π₯οΈ Open it in Cursor
- π€ Ask your Cursor agent:
"Help me implement the AI First Workspace Template for my company. Here's the template: https://github.com/vsevolodustinov/ai-first-workspace-template Guide me through: - Setting up the repository structure - Customizing it for [Your Company Name] - Creating the GitHub repositories - Migrating our team to this framework Walk me through each step and handle the technical setup."π― The AI will handle: Repository creation, script generation, content customization, team onboarding plans, and technical setup.
π You focus on: Strategic decisions, team alignment, and organizational change management.
The manual instructions below are for reference only - let AI handle the implementation!
# 1. Fork or download this template
git clone https://github.com/vsevolodustinov/ai-first-workspace-template.git
cd ai-first-workspace-template
# 2. Create your company's GitHub organization (if not exists)
# Go to https://github.com/organizations/new
# 3. Create separate repositories for each department:Create these repositories in your organization:
Core Department Repos:
[YourOrg]/Strategy- Strategic planning & competitive intelligence[YourOrg]/Product- Product roadmap & specifications[YourOrg]/SalesAndMarketing- Marketing campaigns & sales processes[YourOrg]/Operations- Operational processes & metrics[YourOrg]/Operations-Hiring- Hiring processes & recruitment[YourOrg]/Finance- Financial models & projections[YourOrg]/Legal-HR- Contracts, policies & HR workflows
Technical & Project Repos:
[YourOrg]/Dev-[ProductName]- Your main product codebase[YourOrg]/Projects- Client project portfolio management[YourOrg]/Presales- Presales materials & proposals
Main Workspace Repo:
[YourOrg]/SharedWorkspace- This template, customized for your company
π‘ Naming Your Main Workspace Repository:
You can choose any name for your main workspace repository. Recommended:
SharedWorkspace
SharedWorkspace- Short, practical name for daily use[YourCompany]-Workspace- Include your company nameAI-Workspace- Emphasize the AI-first approachThe name you choose here will be used in git clone commands and daily workflows, so pick something convenient for your team.
# Replace [YourOrg] with your GitHub organization name
# Replace [YourCompany] with your company name
# Create and initialize each repository:
gh repo create [YourOrg]/Strategy --private
gh repo create [YourOrg]/Product --private
gh repo create [YourOrg]/SalesAndMarketing --private
gh repo create [YourOrg]/Operations --private
gh repo create [YourOrg]/Operations-Hiring --private
gh repo create [YourOrg]/Finance --private
gh repo create [YourOrg]/Legal-HR --private
gh repo create [YourOrg]/Dev-[ProductName] --private
gh repo create [YourOrg]/Projects --private
gh repo create [YourOrg]/Presales --private
gh repo create [YourOrg]/SharedWorkspace --private
# Clone and populate template content:
git clone git@github.com:[YourOrg]/SharedWorkspace.git
cd SharedWorkspace
# Copy template content to each repository
# (Detailed instructions in setup/ folder)# Edit .cursorrules to replace [COMPANY_NAME] with your company
# Edit setup scripts to point to your repositories
# Update README.md with your company information
# IMPORTANT: Set up .gitignore for your workspace
cp .gitignore.template .gitignore
# This file excludes department repos (Docs/, Dev/, etc.) since they're separate git repositories# For each department, copy relevant template content:
cd Docs/Strategy
# Copy strategy templates and examples
# Replace Elly Analytics examples with your company information
cd ../Product
# Copy product templates and examples
# Adapt to your product architecture
# Repeat for all departments...# Add team members to appropriate repositories
# Set up department-specific permissions
# Train teams on AI context switching# Install Cursor for all team members
# Import workspace configuration
# Train on AI context switching commands- Department heads - Learn context switching and AI workflows
- Team members - Daily workflow patterns and AI collaboration
- Leadership - Cross-department visibility and strategic AI use
- Week 1-2: Department heads set up repositories and learn workflows
- Week 3-4: Migrate existing documentation to new structure
- Week 5-6: Train team members on AI-first workflows
- Week 7+: Full adoption and optimization
- Leadership Alignment - Ensure executive buy-in for AI-first transformation
- Repository Setup - Follow the Implementation Guide above
- Pilot Department - Start with one department (recommend Strategy or Product)
- Gradual Rollout - Expand to additional departments based on early success
- Training & Support - Invest in team training for AI collaboration workflows
-
Clone your organization's workspace:
git clone git@github.com:[YourOrg]/SharedWorkspace.git cd SharedWorkspace -
Run setup to clone all departments:
# For Unix/Mac: bash setup/clone-all-repos.sh # For Windows: .\setup\clone-all-repos.ps1
-
Start with AI context switching:
"Use strategy context" "Use product context" "Use marketing context"
YourCompany-SharedWorkspace/
βββ .cursorrules # AI context switching & company rules
βββ README.md # This documentation
βββ setup/ # Automated setup scripts
βββ Docs/ # Department repositories (separate repos)
β βββ Strategy/ # Strategic planning & competitive intel
β βββ Product/ # Product roadmap & specifications
β βββ SalesAndMarketing/ # Marketing campaigns & sales process
β βββ Operations/ # Operational processes & metrics
β β βββ Hiring/ # Hiring processes & recruitment
β βββ Finance/ # Financial models & projections
β βββ Legal-HR/ # Legal contracts & HR policies
βββ Dev/ # Technical repositories
β βββ [ProductName]/ # Your product codebases
βββ Projects/ # Client projects (separate repos)
βββ Presales/ # Presales materials (single repo)
π Department Autonomy - Each team owns their repository and workflows π€ AI Context Switching - AI adapts expertise based on department context π Cross-Department Visibility - Transparency while maintaining specialization π Proper Access Control - GitHub-based permissions for enterprise security β‘ Developer-Quality Tools - Professional workflows for all team members
- GitHub Organization - For repository management and access control
- Cursor Editor - Primary AI-enabled development environment
- Git Proficiency - Basic git knowledge for team members
- SSH Keys - Secure repository access (setup guide included)
- macOS - Full support with automated setup scripts
- Windows - Full support with PowerShell scripts
- Linux - Full support with bash scripts
- Leadership Champion - Executive sponsor for AI transformation
- Technical Coordinator - Person to manage repository setup and permissions
- Department Heads - Owners for each department repository
- Training Plan - Structured onboarding for team members
- Setup Guides - Step-by-step implementation instructions
- Workflow Examples - Real-world usage patterns from each department
- AI Context Rules - Complete prompt engineering and behavior definitions
- Template Content - Professional examples for all department types
- Issues - Report problems or request features via GitHub Issues
- Discussions - Share experiences and get community support
- Examples - Extensive real-world examples from Elly Analytics
Organizations using this framework typically see:
- 30-50% reduction in routine administrative work
- Improved cross-department collaboration through AI-enabled transparency
- Faster onboarding for new team members
- Higher quality documentation through AI assistance
- Better strategic alignment across all departments
This template improves through real-world usage. If your organization adopts this framework:
- Share feedback on what works and what doesn't
- Contribute improvements via pull requests
- Document adaptations for different industries or team sizes
- Help others through discussions and issue responses
MIT License - Feel free to adapt this framework for your organization's needs.
Created by Seva Ustinov based on the real-world implementation at Elly Analytics.
Special thanks to the Elly Analytics team for pioneering AI-first organizational workflows and proving this approach works at scale.