Divya

Posted on Jan 24 • Edited on Feb 15

AI Based 24*7 Study Guide 👩🏻‍💻📓✍🏻💡

#devchallenge #agentaichallenge #ai #machinelearning

This is a submission for the Agent.ai Challenge: Assembly of Agents (See Details)

What I Built

I created a Study Guide Agent to make learning smarter, easier, and more engaging. Users can input a topic, choose their education level or purpose (exams, academics, research), and get a detailed explanation tailored to their needs. To help retain information, the agent generates fun stories, and learning tips customized for better understanding and memory.

It doesn’t stop there 😁! Users can create a custom quiz to test their knowledge. The agent assesses their answers, providing personalized feedback, model answers, and actionable learning suggestions for improvement. This all-in-one guide is designed to simplify learning and make it enjoyable!

I built this because I had long wanted to, and this is closely related to a project I built this December as well:- AI-Study-Guide

Demo

My agent is live at:- Study Guide

A video demo showcasing my working agent:-

Agent.ai Experience

🎉 So, here it is— my 3rd agent on the agent.ai platform! 🎉

This one took a lot more planning compared to my first, but guess what? It’s totally worth it! ☺️😁

The process boils down to these key steps:

1️⃣ Start with the basics: Collect the user's education level and the topic they want to learn or revise.

2️⃣ Generate detailed explanations: Use an LLM agent to create a comprehensive, well-structured explanation of the topic tailored to the user’s needs.

3️⃣ Add a storytelling twist: Ask users if they’d like a story to make the topic memorable.

4️⃣ Bring in the Story Maker: Call the Story Maker Agent to craft a fun, engaging story based on the topic to help with retention.

5️⃣ Boost learning: Offer learning tips to simplify and reinforce the topic.

6️⃣ Create custom learning aids: Use an LLM agent to generate some learning tips, ensuring they’re tailored for the user as per the topic.

7️⃣ Quiz Questions Count: Let users pick how many quiz questions they’d like to assess their grasp of the topic.

8️⃣ Quiz time: Use an LLM agent to create a subjective quiz and present it to the user.

9️⃣ User input: Provide a text area for users to type their answers for each question based on their understanding.

🔟 Feedback and growth: Call an LLM agent to evaluate their answers, provide personalized feedback, model answers, as well as actionable suggestions for improvement.

🔢 Final output: Present everything back to the user— feedback, suggestions, and model answers— so they can deepen their understanding and grow in their learning journey.

Here’s a sneak peek of the actions tab of my agent, showing the workflow behind it all: 👇

Conclusion

This marks my third project on the Agent.ai platform, my second requiring interaction with another agent, and the most extensive one in terms of planning and execution. While I acknowledge it’s not the most time-efficient— since all content is generated afresh every time the agent is run, even if users don’t need the story or tips— I still like it, for now. I plan to refine and optimize it post my semester exams (because, let’s face it, there’s so much to do and barely a week left). Stay tuned for updates! 🚀

So, that's it!

If you're still here, Thank you ☺️🙏🏻.

Please share your thoughts, feedback, and suggestions in the comment section below 🗨️

GIF showing a studying hard cute penguin

What is MCP? No, Really!

See MCP in action and explore how MCP decouples agents from servers, allowing for seamless integration with cloud-based resources and remote functionality.

Watch the demo

Top comments (14)

Fayaz • Jan 30

Nicely done!
Congrats on winning! ❤️

Divya • Jan 31

Thank you
It really was a pleasant surprise for me 😁

dummy • Jan 24

It's a great one!
There's always room for improvement - It's all part of the process, so don’t stress too much about it. Once the main part is done, the rest will follow naturally.

And you’ll rock in your sem exams..! All the very best..