Complete Workflow Guide

Build a Software Development Team in 10 Minutes

Create a fully functional AI-powered software development team (Developer, QA Engineer, DevOps Engineer) with automated optimization, multi-agent orchestration, and production-ready deployment.

10 Minutes
3-Agent Team
Optimized Performance

What You'll Build in 10 Minutes

Development Team
Developer, QA Engineer, DevOps Engineer
GEPA Optimization
Significant performance improvements
Multi-Agent Orchestra
Coordinated team workflows
Production Ready
Real-world deployments

Requirements

Hardware

  • 16GB GPU RAM for optimization (Step 5)
  • 8GB+ System RAM recommended
  • Stable internet for model downloads

Software

  • Python 3.11+
  • Git (verify: git --version)
  • Ollama (optional, for local LLMs)
Windows Users

Set PYTHONUTF8=1 environment variable:

set PYTHONUTF8=1

Installation

Choose your preferred installation method

Recommended: uv (Fastest)

This guide uses DSPy for GEPA optimization and orchestration.

1. Install uv:

curl -LsSf https://astral.sh/uv/install.sh | sh

2. Install SuperOptiX with DSPy:

uv pip install -U superoptix[frameworks-dspy]

3. Verify installation:

super --help
Two Ways to Use SuperOptiX

Choose Your Interface

After cd swe, you can use either Super CLI or traditional commands

BETA

💬 Super CLI

Talk in Natural Language

Conversational
Just describe what you want
Beginner Friendly
No commands to memorize
AI-Powered
Understands intent automatically
$ super
SuperOptiX › build a developer agent
✨ Creating your agent...

⚙️ Traditional CLI

Precise Commands

Full Control
Explicit commands and flags
Scriptable
Perfect for automation
Advanced Features
Access to all options
$ super agent pull developer
$ super agent compile developer
✅ Agent compiled successfully

💡 Pro Tip: Start with Super CLI to learn, then switch to traditional CLI for automation and scripting

The Complete Workflow

Follow these 10 steps to build your AI development team (using traditional CLI commands)

1

Initialize Your Project

Create your software engineering project structure.

super init swe
cd swe

Created structure: agents/, evals/, guardrails/, knowledge/, memory/, optimizers/, orchestras/, pipelines/, protocols/, servers/, teams/, tools/

2

Pull a Developer Agent

Download a pre-built developer agent from the marketplace.

super agent pull developer

Downloaded: agents/developer_playbook.yaml with complete SuperSpec DSL configuration

3

Compile the Agent

Generate executable Python code from the SuperSpec DSL.

super agent compile developer

What happens: Your agent specification is compiled into executable code with built-in evaluation scenarios, ready to run in pipelines/developer_pipeline.py

4

Evaluate the Agent

Test against RSpec-style BDD scenarios to establish baseline performance.

super agent evaluate developer

💡 Don't worry about initial scores!

Agents start with generic prompts. Optimization in Step 5 will significantly improve performance!

5

Optimize with GEPA (The Magic Step!)

GEPA automatically improves agent instructions through evolutionary optimization!

super agent optimize developer --auto medium

⚡ What happens during optimization:

  • Analysis: GEPA analyzes failed scenarios
  • Mutation: Generates improved prompt variations
  • Evaluation: Tests each variation
  • Selection: Keeps best-performing prompts
  • Iteration: Repeats until convergence
Light
5-10 min
Medium ⭐
15-25 min
Intensive
30-60 min
6

Re-Evaluate After Optimization

Load optimized agent and measure improvement.

super agent evaluate developer --load-optimized

🏆 Significant performance improvements!

GEPA optimization dramatically increases agent accuracy and reliability

7

Run the Agent

Test your optimized agent on a real-world task.

super agent run developer --goal "Implement a user registration API endpoint with email validation"
8

Add More Agents (Build Your Team!)

Pull QA Engineer and DevOps Engineer to complete the team.

# Pull QA engineer
super agent pull qa_engineer

# Pull DevOps engineer
super agent pull devops_engineer

# Compile both agents
super agent compile qa_engineer
super agent compile devops_engineer

💡 Pro Tip: You can optimize these agents too!

For this quick start, we'll use them as-is for the orchestra. But feel free to run optimize + evaluate on them for even better results!

✅ You now have a complete 3-agent team ready for orchestration!

9

Create & Run Multi-Agent Orchestra! 🎭

The Kubernetes moment! Orchestrate your agents like a container platform.

1. Create the Orchestra:

super orchestra create sdlc

Auto-detects your agents and creates orchestras/sdlc_orchestra.yaml

2. List Available Orchestras:

super orchestra list

3. Run the Orchestra:

super orchestra run sdlc --goal "Build a complete web application for a task management system with user authentication, CRUD operations, and real-time notifications. Include comprehensive testing and deployment configuration."

🎭 What Just Happened (The Magic!):

1

Developer analyzed the goal and provided implementation reasoning

2

DevOps Engineer created CI/CD pipeline based on developer's output

3

QA Engineer created comprehensive test plan based on both outputs

The orchestra automatically:

  • Coordinated agent communication
  • Passed context between tasks
  • Generated artifacts in workspace
  • Completed full SDLC workflow

📋 Generated Artifacts:

  • • implement_feature_implementation.txt
  • • configure_ci_pipeline_result.json
  • • create_test_plan_test_plan.txt

🎉 The Kubernetes Moment!

SuperOptiX just orchestrated your agents like a container orchestration platform:

Orchestrated automatically - Like Kubernetes manages containers
Managed communication - Seamless data flow
Coordinated responses - Each built on previous
Generated artifacts - Production-ready
10

Observe & Monitor

Inspect agent traces and monitor performance.

View traces:

super observe traces developer

Launch dashboard:

super observe dashboard
Framework-Agnostic by Design

Multi-Framework Support

SuperOptiX comes with DSPy and GEPA by default, giving you the most powerful optimization out of the box.

But you're not locked in! Use the exact same workflow with any of 6+ frameworks:

DSPy (Default)
OpenAI SDK
CrewAI
Google ADK
Microsoft
DeepAgents
🔷
Default

DSPy

Stanford research framework with powerful optimization capabilities

Best for:

  • Complex reasoning tasks
  • Research applications
  • Advanced optimization

Quick start:

pip install -U superoptix[frameworks-dspy]
super agent pull sentiment_analyzer
🤖

OpenAI SDK

Simple agents with OpenAI's powerful models

Best for:

  • Simple agents
  • Fast prototyping
  • OpenAI ecosystem

Quick start:

pip install -U superoptix[frameworks-openai]
super agent pull assistant_openai

CrewAI

Multi-agent collaboration

Best for:

  • • Multi-agent teams
  • • Role-based agents
  • • Collaborative workflows

Install:

pip install -U superoptix[frameworks-crewai]

Pull agent:

super agent pull researcher_crew

⚠️ Cannot be installed with DSPy (json-repair conflict)

🔮

Google ADK

Gemini 2.0 native

Best for:

  • • Gemini integration
  • • Free tier available
  • • Google ecosystem

Install:

pip install -U superoptix[frameworks-google]

Pull agent:

super agent pull assistant_adk

Setup API key:

export GOOGLE_API_KEY=your-google-api-key
💠

Microsoft

Enterprise Azure

Best for:

  • • Enterprise Azure
  • • Microsoft ecosystem
  • • Corporate deployments

Install:

pip install -U superoptix[frameworks-microsoft]

Pull agent:

super agent pull assistant_microsoft
🌊

DeepAgents

Complex planning

Best for:

  • • LangGraph planning
  • • Advanced reasoning
  • • Complex workflows

Install:

pip install -U superoptix[frameworks-deepagents]

Pull agent:

super agent pull research_agent_deepagents

⚡ GEPA: The Universal Optimizer

Same command works on ALL 6 frameworks! This is the revolutionary power of SuperOptiX.

For ANY Framework:

# Works on DSPy, OpenAI SDK, CrewAI, Google ADK, Microsoft, DeepAgents!
super agent optimize <agent_name> --auto medium

# Examples:
super agent optimize sentiment_analyzer --auto medium        # DSPy
super agent optimize assistant_openai --auto medium          # OpenAI
super agent optimize researcher_crew --auto medium           # CrewAI
super agent optimize assistant_adk --auto medium             # Google ADK
super agent optimize assistant_microsoft --auto medium       # Microsoft
super agent optimize research_agent_deepagents --auto medium # DeepAgents

Optimization Levels:

Light~5 min

Quick iteration, prototyping (2-3 iterations)

Medium ⭐ Recommended~10-15 min

Most use cases (5 iterations)

Intensive~30+ min

Critical production agents (10+ iterations)

✅ Framework-Agnostic Optimization

No framework-specific tricks. No vendor lock-in. Just pure, portable optimization across your entire agentic stack.

🔬 Try Different Frameworks - Complete Examples

Complete OpenAI SDK Workflow:

# 1. Initialize
super init openai_project
cd openai_project

# 2. Pull OpenAI agent
super agent pull assistant_openai

# 3. Compile
super agent compile assistant_openai

# 4. Baseline evaluation
super agent evaluate assistant_openai
# Result: Check initial performance

# 5. Optimize with GEPA
super agent optimize assistant_openai --auto medium
# GEPA optimizes agent

# 6. Re-evaluate
super agent evaluate assistant_openai --load-optimized
# Result: 100% pass rate!

# 7. Run
super agent run assistant_openai

🎉 Congratulations!

You've Explored the Future of Agent Development!

You just experienced a framework-agnostic agent platform with optimization at its core

What You Just Accomplished:

  • Built a complete AI team

    Developer, QA, DevOps working together

  • Optimized with GEPA

    Automatic performance improvements

  • Tested with BDD scenarios

    Production-grade validation

  • Explored 6+ frameworks

    DSPy, OpenAI, CrewAI, Google, Microsoft, DeepAgents

  • Orchestrated multi-agent workflows

    Production-ready artifacts generated

The Big Picture:

🔓 Zero Lock-In

Switch frameworks anytime. Your agents, optimizations, and tests stay portable.

⚡ Optimization-First

GEPA works universally across all frameworks. Same command, consistent improvements.

🎯 Production-Ready

Built-in evaluation, orchestration, and monitoring from day one.

What Makes SuperOptiX Unique:

🔧

Framework-Agnostic

First platform that works universally across all major agent frameworks

🧬

Optimization at Core

GEPA optimizer built-in from the start, not an afterthought

🎼

Multi-Agent Native

Orchestra features for seamless agent coordination

🚀 Next Steps:

  • Try Genies tier agents with tools & RAG
  • Browse 165+ pre-built agents in marketplace
  • Create custom agents with SuperSpec DSL
  • Build complex multi-agent organizations
  • Deploy to production with observability