Entertainment Assistant

Inspiration

There have been so many times when we have sat down, either with family or by ourselves, looking for something good to watch, but cannot find anything interesting. So much time is wasted not actually watching, but looking for something to watch. That's how we came up with the idea for an entertainment assistant, well-rated, popular recommendations based on our interests and access to streaming services.

What it does

This assistant takes in several variables and asks the user questions to gauge its interests. It then provides 10 recommendations for movies you might like based on what you are looking for, and will give summaries, more recommendations and add additional variables until you find what you want to watch.

How we built it

We built it using VoiceFlow's platform, using prompt chaining and GPT-4

Challenges we ran into

Neither of us have ever used VoiceFlow before or made a project with an LLM. Learning how VoiceFlow works was a little tedious but once we figured it out, we were well on our way. Another challenge was learning how to customize and troubleshoot the intents in VoiceFlow. Occasionally, training the chatbot would produce weird loops in which intents were triggered without having any of the intent criteria met.

Accomplishments that we're proud of

We are proud that we were able to accomplish a working chatbot, given this weekend has been especially busy and we were unable to spend as much time working on it as we might have hoped for.

What we learned

We have learned a lot about using LLM's and VoiceFlow's platform, and we hope to continue to use it in the future

What's next for Entertainment Assistant

Entertainment Assistant might be made into a working app, where users can have visuals, a chatbot, and more features to find what they might enjoy watching.

Built With

• gpt
• voiceflow