FutureSmart Agent
A complete AI Agent system for real use cases. RAG, NL2SQL, API actions, multi-agent logic — everything in one place.
Read more
K
Votes0
Score0
🧠Community0×5= 0
🌐Public0×1= 0
Karthik Sathish
Think Less
Individual
May 30, 2025 at 05:30 AM
Claude AI. Cartesia (for creating the interviewer's voice in the demo)
hear-about-us
name
Karthik Sathish
linkedIn-profile
Project Title
Think Less
Describe what your project does?
This project helps users during online meetings by answering questions asked by others in real-time. Before the meeting starts, the user uploads a set of documents, and the system uses those to provide answers based on the content. It also generates a summary of the meeting so the user can review what was discussed later.
The underlying architecture: One of the key things I focused on was efficiency. Instead of sending the entire meeting transcript to the language model, I only send the actual question. This has two big advantages:
The answer is more relevant since it’s based just on the question. It keeps token usage low, which helps reduce costs.
To detect questions, the system looks for sentences that start with WH-words like what, where, how and so on. If it finds one, it treats it as a question and sends it to the language model for a response.
What AI tools did you use?
Claude AI. Cartesia (for creating the interviewer's voice in the demo)
Share a 3–5 minute demo video (Google Drive link only) explaining your core idea, tools used, challenges faced, and how you improved your solution — make sure the link is viewable.
Is this project your original and proprietary work?
Yes
Does it include any private, sensitive, or restricted code or content?
No
Do you give us (AI Demos) permission to feature your work on our website and share it with our community (with full credit)?
Yes
Do you consent to us showcasing your name/profile on our social media platforms and YouTube
Yes
Connect with Karthik Sathish to learn more about their project.