Lapidador

Project summary

Context

Proprietary product created from a pain point that kept showing up in class: people understood prompt theory but got stuck when trying to turn a real need into a good instruction.

Business problem

Students were often overwhelmed by the number of prompting techniques available, which made learning harder by making the topic seem far more difficult than it really is. There was no way to practice with guidance, receive a ready prompt, and still understand how the AI produced the result.

Audience & personas

• Business professionals who use AI at work
• Students in AI courses and workshops
• People who need better prompts without learning the technical taxonomy

Adoption barriers

• Difficulty explaining context to the AI
• Prompts that are too generic
• Low autonomy after training

Solution design

• Conversational flow that asks for context when information is missing
• Catalog of 23 techniques applied by the system in a personalized way
• Credit system with Mercado Pago payment integration
• Final response with a ready prompt and instructional explanation

Assets created

• Technique catalog
• Prompt structure recommendation engine
• Landing page, authenticated application, and payment flow

Metrics & signals

• 23 techniques integrated
• Product developed and published independently in just two weeks
• Full stack in React, TypeScript, Express, and AWS

Lessons learned

Users learn more when the tool shows the reasoning, not just the final answer.

Next improvements

• Add examples by role, version history, and a corporate mode with a team prompt library.
• Build an MCP server to integrate Lapidador's engine directly into chatbot tools.