Hackathon prototype & AI-assisted campaign planning

AdDNA Campaign Prototype

A one-day prototype turning a company URL into an editable starting point for an advertising campaign.

View the workflow Explore the build

Business problem

The challenge

Campaign planning required repeated website research and manual decisions about audiences, keywords and budgets.

Technical solution

Our approach

Next.js frontend, Java Spring Boot REST backend and AI-assisted website analysis.

  1. 01

    Bootstrap the prototype

    Bootstrapped frontend and backend for a ten-person team spanning product, database and engineering. Learned Java during implementation; configured Next.js, React, TypeScript, Tailwind, Spring Boot, PostgreSQL and Turborepo.

  2. 02

    Turn a URL into editable recommendations

    Built a workflow for product summaries, age/gender/income/interests, audience keywords, budget amounts and campaign duration, with editable fields. Added Audience Mood Scanner, trending keywords and sentiment presentation.

  3. 03

    Explore the campaign-planning scope

    The initial concept also included scraping/RAG, location/device targeting, budget dates and creative-image inspiration. The prototype includes mock fallbacks and randomized estimates; those broader outputs need further validation.

  4. 04

    Hand off the next stage

    Teammates subsequently developed the production Shopper Graph/MNTN Match system. Its scraping, LangChain/OpenAI generation and taxonomy retrieval are later team work.

Results & impact

Project outcomes

A one-day prototype turning a company URL into an editable starting point for an advertising campaign.

One day

Working hackathon prototype

A cross-functional ten-person team produced an end-to-end demo.

Editable

Campaign-planning starting point

Users could revise generated summaries, demographics, keywords and budget/duration recommendations.

Visual assets

How the workflow fits together

AdDNA Campaign Prototype

flowchart TD
  A["Company URL"] --> B["Java analysis API"]
  B --> C["AI-assisted analysis"]
  C --> D["Summary, audience and budget"]
  D --> E["Editable planning UI"]
01 Next.js frontend, Java Spring Boot REST backend and AI-assisted website analysis.

Tech stack

Technology and ownership.

Hackathon prototype ownership covers the application foundation and documented UI contributions. Later production Shopper Graph work belongs to the subsequent team. Earlier portfolio notes associate the broader initiative with approximately 15% reductions in build time/churn; prototype attribution and measurement remain unconfirmed.

Next.jsReactTypeScriptTailwind CSSJavaSpring BootAI APIsPostgreSQLTurborepo