NASA Bidara prompt engineering techniques

Although ChatGPT, or like LLM models using natural language as interface, works great for many use cases. However, to achieve a desired result for specialized use cases without coding we still need to apply some good prompt engineering strategies.

In a nutshell, prompt engineering is a technique that blends both business and technical understanding — which involves creativity and strategic thinking.

Why the Quality of the Prompt Matters?

While interactions with large language models (LLMs) are natural, the quality of prompts has a crucial role in how they perform reasoning and response generation.

For instance, NASA’s innovative approach use on creation of Bidara bot involves crafting system-prompt designs that trigger LLM networks to generate high-quality responses for users while meeting specific constraints set by both businesses and engineers. See here to learn more about Bidara (https://nasa-petal.github.io/bidara-deep-chat).

Figure: NASA Inspired Prompt Engineering

What is NASA Bidari Chat Bot?

BIDARA is a GPT-4 chatbot that was instructed to help scientists and engineers understand, learn from, and emulate the strategies used by living things to create sustainable designs and technologies.

Bidara chatbot guide users through the Biomimicry Institute’s Design Process, a step-by-step method to propose biomimetic solutions to challenges. This process includes defining the problem, biologizing the challenge, discovering natural models, abstracting design strategies, and emulating nature’s lessons.

Objectives

• The objective of this writing is to examine a “Good Prompt Design” — specifically examine NASA famous Bidara chatbot — system prompt.
• Write a Model file in Ollama to create a local version of Nasa-Bidara
• Writing prompt in structure way
• Look at some prompt augmenting tools without writing prompts (Future Writing)

How it Works?

let’s try it out first by re-creating NASA Bidara Chatbot in Ollama

Step-1. Setup Local Ollama / Open WebUI