Prompts to Product — whats best practice for deploying prompts in indie / low-code development?

Question

Hey builders 🛠️
Anyone out there getting their prompts into products? I've been creating tools for myself with some prompt engineering. Now I want to create them into a standalone tool.

I would like to use some low-code or no-code tools but I haven't found much info or best practices for implementing with those kind of platforms and managing the prompts.

What does your stack, workflow or practices look like for going from prompt crafting, implementation and into a shipped product and ongoing management?

Andres Rosas Sanchez · Accepted Answer

Hi Owen, At Neosmart, we have a prompting tool that we use to optimize and collect our most powerful and useful prompts. This tool is also connected to ChatGPT-4, so you can perform your tasks within the proper platform. Please check it out and let me know what you think. We will be launching it soon. https://neoprompts.ai/en

Answer

I've been using GPT-3 and GPT-4 via OpenAI's API to generate content and enhance my prompts. For a no-code platform, you could try Bubble which has plugins for OpenAI and Anthropic. Streamlit is another option if you're comfortable with a little Python - super easy to create web apps and integrate LLMs. Key is to have a clear prompt structure, do output parsing if needed, and have human oversight. Interested to see what stack others are using for prompt-to-product!

Prompts to Product — whats best practice for deploying prompts in indie / low-code development?

Replies