Skip to content

KingasRahd/YTIntel

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

📺 YTIntel

YTIntel is an AI-powered YouTube channel analyzer that helps users find the most valuable videos tailored to their personal learning goals. Instead of scrolling endlessly, YTIntel delivers a curated list of content that aligns with your intent, not just the algorithm.


Live Demo

Link : https://ytintel.streamlit.app/


🚨 Problem

YouTube is overflowing with valuable content — but when a user wants to learn something specific, finding the right videos becomes a frustrating mess.

Search results are noisy, clickbait-heavy, and popularity-driven

Channel pages contain hundreds of videos with little guidance

Even playlists, when available, are generic and not personalized

Users waste time scrolling, guessing, and getting distracted, especially when they’re just trying to follow a goal like “learn AI for research” or “build confidence”

For learners and upskillers, this is a huge productivity leak.


✅ How YTIntel Solves It

YTIntel is an intelligent YouTube channel summarizer that helps users find only the most relevant videos from any channel — based on their personal goals.

Using:

🎯 User-defined goals

📺 Channel video metadata (title, description, tags, etc.)

🤖 Gemini LLM via LangChain for deep filtering

📊 Relevance scoring and justification-based ranking

YTIntel:

Picks only the videos that align with the user’s intent

Categorizes them meaningfully (e.g., "Foundational", "Actionable Advice", etc.)

Provides brief explanations for why each video was chosen

Filters out everything else

In short:

YTIntel turns chaotic video libraries into goal-focused learning paths — instantly. No playlists. No guesswork. No fluff.


🚀 Features

  • 🔎 Channel Summarization – Quickly get a summary of the channel: name, subscribers, video count, etc.
  • 🎯 Goal-Based Filtering – Users enter their learning goal, and YTIntel finds videos that truly match it.
  • 🧠 LLM-Powered Analysis – Uses Gemini to deeply understand video titles, descriptions, and metadata.
  • 📊 Streamlit UI – Clean, simple, interactive interface.
  • 📷 Video Thumbnails + Links – Beautiful video display in YouTube style.
  • Categorizes the video-It categorizes the video for easier navigation

🛠️ Tech Stack

Tech Purpose
Python Core logic & backend
Streamlit Frontend / App UI
YouTube Data API v3 Fetch channel & video metadata
Gemini LLM (via LangChain) Video understanding and filtering
Pandas Data wrangling
isodate / datetime Duration parsing

✍️ How It Works

  1. User Input:

    • Enters a YouTube channel ID
    • Describes their learning goals
  2. Video Extraction:

    • Title, description, duration, likes, views, etc. are fetched via the YouTube API
  3. LLM Filtering:

    • Gemini filters and analyzes videos to find those that are most relevant to the user's goals
  4. Output:

    • Displays the selected video cards with linked thumbnails,titles, and justifications and category

🧪 Sample Use Cases

  • AI/ML Enthusiast: "I want to learn core machine learning concepts for research."
  • Self-Help Seeker: "I want to build better habits and develop discipline."
  • Student: "I'm preparing for data structures & algorithms interviews."

🧠 YTIntel adapts to each intent and shows only what's useful—no clickbait, no fluff.


💖 Credits Built by @Sagnik

Powered by Google's Gemini, YouTube Data API, and ✨ Streamlit magic


🙌 Star the repo if you found it helpful!

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages