Skip to content

the-dress-code/talk-prep

Repository files navigation

TalkPrep

Why does TalkPrep exist?

This project is / was an experiment.

I asked Claude to build me a 2-3 week plan to level up in my agentic engineering skills. "We" worked on building a tool called talk-prep, and the building of that tool would be part of my leveling-up plan.

My preliminary vision for talk-prep:

Assist me in organizing my ideas into usable abstracts and outlines for future conference speaking opportunities, with some AI-powered features. Don't write the talk for me. Don't summarize what I've said. Help me organize my own thoughts about my own topic(s) and give me back me own words to me in a structured manner to help me build or develop an abstract, outline, etc. Show me where I make claims that need further support.

So far, the most meaningful thing the tool does is take what I call braindump text - your notes and thoughts about your topic - and return structured JSON : with topic / point / details. This action is achieved with the braindump_processor using Qwen 2.5 via Ollama.

WIP: socratic_questioner will surface claims from your braindump test and ask you to support your claim if your provided evidence is lacking.

Installation

If available in Hex, the package can be installed by adding talk_prep to your list of dependencies in mix.exs:

def deps do
  [
    {:talk_prep, "~> 0.1.0"}
  ]
end

Documentation can be generated with ExDoc and published on HexDocs. Once published, the docs can be found at https://hexdocs.pm/talk_prep.

About

Let Qwen 2.5 via Ollama help turn your thoughts into talks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors