Lorenzo — AI Campaign Strategy Builder

link https://drive.google.com/file/d/1ZIvLmUBAwKuz2etdoRIEhRgkyS9SR5As/view?usp=sharing

1. Extracted Text Content & Summary Based on the video's audio and on-screen text, here is the core sequence of events:

Introduction: You (Brehan) introduce a platform called Lorenzo, built for a smart mobile advertising company.

The Feature: You demonstrate the "Strategy Builder," a tool for creating new advertising campaigns.

AI Chat Interface: Instead of traditional forms, the platform uses an LLM-powered chat interface to gather campaign details.

Intelligent Validation: You input a product name ("Football") and a nonsense description ("bla bla bla"). The LLM correctly analyzes the input, realizes it doesn't match the product, and asks for a more descriptive, relevant input.

Contextual Understanding: Once provided with a valid description (a campaign about American Football in Europe/Premier League), the system accepts it and moves forward.

Dynamic Suggestions: Based on the campaign description and selected industry (Sports > American Football), the LLM automatically suggests optimal targeting metrics, including:

Age Ranges: Young Adults (18-24), Early Adults (25-34), Mid Adults (35-44).

Interest Groups/Keywords: Football Enthusiasts, Premier League, Sports Fans, Match Viewers, etc.

Technical Architecture (Mentioned at the end): You note that your work focused on integration, AWS Lambda API deployment, and the underlying LLM scripts.

2. Key Portfolio Takeaways (What to Highlight) You can use these points directly in your resume or portfolio write-up to showcase your specific skills and the business value of the project.

Project Title: AI-Powered Campaign Strategy Builder (Lorenzo Platform)

Your Role/Focus: Backend & AI Integration (LLM scripting, API deployment)

Key Features to Highlight:

Conversational UI: Replaced static web forms with an intuitive, LLM-driven chat interface for campaign onboarding.

Automated Input Validation: Engineered LLM prompts to analyze user inputs in real-time, successfully identifying and rejecting nonsensical or irrelevant data to ensure high-quality data collection.

Predictive Targeting: Utilized LLMs to automatically generate intelligent audience targeting suggestions (age ranges, specific interest groups) based on brief user descriptions.

Tech Stack Mentioned: LLMs (Large Language Models), AWS Lambda, API Deployment, Backend Integration.