AI Engineering

Project’s dates

Start Date - 16/05/2025
Estimatted End Date - 04/07/2025
Actual End Date - [Actual End Date]

Project Idea Definition

[Describe your project idea in a few sentences]

The AI Packager is an innovative AI-driven automation tool designed to streamline the software packaging process for enterprises. It allows IT administrators to effortlessly create validated deployment packages using intelligent agents, the PowerShell App Deployment Toolkit (PSADT), and automated VM testing. Leveraging advanced retrieval methods and AI-generated scripting, this project drastically reduces manual packaging efforts, ensuring high-quality, ready-to-deploy software installations every time.

🧱 Architecture Overview (MVP)

plaintext
KopierenBearbeiten
                ┌────────────────────────────┐
                │         Flask UI           │
                │ - Upload installer         │
                │ - View generated script    │
                └────────────┬───────────────┘
                             │
                ┌────────────▼────────────┐
                │     Orchestrator Agent  │  <-- Central coordinator
                │ - Routes task flow      │
                └────────────┬────────────┘
                             │
          ┌──────────────────┴────────────────────┐
          │                                       │
 ┌────────▼────────┐                     ┌────────▼─────────┐
 │  Script Agent   │                     │  Database Layer  │
 │ (PSADT logic)   │                     │ - Script storage │
 │ - AI Generation │                     │ - Metadata       │
 └─────────────────┘                     └──────────────────┘

Note: Hyper-V Test Agent is excluded in MVP but planned in long-term vision.

🎯 MVP Functionality Checklist

Module	Required for MVP	Notes
Flask UI	✅	Upload installer, display results
AI Script Generator	✅	Generate PSADT from installer details
Orchestrator Agent	✅	Manage tasks between agents
Database (SQLite)	✅	Store generated scripts
Hyper-V Integration	❌ (Later)	Skip until post-MVP
Cloud / On-Prem Infra	❌ (Later)	MVP runs locally

🚀 Sprint Roadmap (4 Weeks Suggestion)

	Week	Focus Area	Output Milestones
done	1	Project Setup & UI	Flask skeleton, upload form, folder structure
in Progress	2	AI Integration & Orchestration	Script Agent working with mock input
	3	Full MVP Integration	Orchestrator links UI → Agent → Output
	4	DB Storage, Testing, Polish	Save results, bugfixes, README, UX touchups

📋 Detailed Task Plan

🧩 1. Project Bootstrap

~~Setup Python venv~~

~~Install Flask~~

~~Create base folder structure:~~

bash
KopierenBearbeiten
/app
  /agents
  /routes
  /templates
  /static
  /models

~~Add .gitignore, requirements.txt, and README.md~~

🧠 2. Flask App Core

~~Upload form: accept .exe or .msi~~

~~Route /upload handles form → saves file~~

Route /result displays AI-generated script

Minimal Bootstrap UI

🤖 3. Orchestration Layer (Core MVP Logic)

Build orchestrator.py

Define interface: generate_script(installer_path: str) -> str

Orchestrator calls Script Agent

🛠️ 4. Script Agent (LLM-Based)

Design input format for agent (installer type, filename, metadata)

Integrate with OpenAI or local LLM (via API)

Prompt engineering for PSADT script style

Return generated script

💾 5. Database Layer (SQLite)

Save installer filename, generated script, timestamp

Create simple DB schema

Add view page to browse history

🧪 6. Testing & Polish

Manual tests (happy path, invalid file, timeout)

Add script download button

Polish UI (titles, spacing)

README with install + usage

Github repo

[Enter github repo link]

GitHub - alexandergreif/AIPackager

Contribute to alexandergreif/AIPackager development by creating an account on GitHub.

github.com

Database Schema

[Enter screenshot of your db shema]

Comparison Table

[Update the comparison table]

Comparison Table

Project Video Recording

[3-5 minutes video recording of your project]

Project Slides

[3-5 slides presenting your project]

Unnamed

[Optional] Screenshots

[Enter screenshots of your project]

Main Topics of Learning Materials

Fundamentals of AI, Machine Learning, LLM’s

Evolution of AI - History of Machine Learning
What changed with LLM’s - ChatGPT Revolution
Current market leads and differences between them

Natural Language Processing

Data Preprocessing
Data Cleaning

API Integration with AI models

Structured Output

Prompt Engineering
GenAI Ethics
Advanced GenAI

Retrieval Augmented Generation (RAG)

External Optional Resources

Video Resources Written Resources

Examples of projects

‣

AI Data Analyst

‣

Ask me anything about a big project

‣

Online Service - Article AI assistant

‣

A tool that uses AI to automatically label large datasets

‣

Image Generation for Ads

Student Project

RagLangBot Project Overview

Examples of bad projects

‣

What should I make for dinner?

‣

Text Auto-completion Tool

‣

Random Image Generator

V1 Milestones

Milestone	Description	Achieved
BA_1		🏅
DB_1
AI_1

[Add Milestones according to your project needs]

V2 Milestones

Milestone	Description	Achieved
BA_X		🏅
DB_X
AI_X

[Add Milestones according to your project needs]

Progress tracking timeline

‣

Week 1 - Project idea definition

‣

Week 2-3 - Backend & Database setup

‣

Week 4 - First GenAI request

‣

Week 5-8 - GenAI iterations

‣

Week 9-10 - [opt] RAG & Extras

‣

Week 11-12 - Finalizing project & Presentation

AI Engineering interview questions

@Alex

AI Engineering - Alex