Deployed Movie Recommendation App with Machine Learning

Every chunking library handles text pretty well. Tables though? They butcher them. HTML tables get split mid-row, columns lose context, and downstream models hallucinate because the structure is gone. We've heard this from teams at every scale. Today we're shipping v1.6 of our chunking library. Tables are now a first-class citizen. Here's what's in the release: 🔹 HTML Table Chunking New TableChef + TableChunker components that extract, normalize, and chunk tabular data while preserving row and column semantics. No more blind token-boundary splits through your structured data. 🔹 Self-Hostable Chunking API Run chonkie serve and you get a full FastAPI-powered chunking server. One command. No auth setup, no billing. Hit it from any language, any service in your infra. Ideal for polyglot stacks or teams that want to centralize all infrastructure. 🔹 Native Async Support Every chunking method now has an async equivalent. No more blocking your event loop. Long-requested, finally here. Finally, a note of thanks. Our python library now gets 120K+ downloads every week. Grateful to everyone who has contributed and continues to support our growth. 🦛 ❤️ #opensource #rag #llm #chunking #ai

The biggest update yet to our OSS library. Every AI builder knows the pain of working with tables. v1.6 solves it. Check it out!

I am excited to share my recent project: RAG-Based Question Answering System. This project is an AI-powered question answering system that retrieves relevant information from web-scraped content and generates accurate responses using a Retrieval-Augmented Generation (RAG) architecture. The system collects data from websites, processes the text into embeddings, stores them in a vector database, and retrieves the most relevant context to generate answers using a Large Language Model. Project Overview The workflow of the system includes: Web scraping to collect information from websites Text cleaning and chunking for efficient processing Converting text chunks into vector embeddingsStoring embeddings in a vector database Performing semantic similarity search to retrieve relevant information Generating context-aware responses using an LLM Delivering the answer through a web interface Technology Stack Python FastAPI FAISS Vector Database Sentence Transformers BeautifulSoup and Requests for web scraping LLM Integration HTML and Jinja Templates for the frontend This project helped me gain practical experience in building end-to-end AI systems including Retrieval-Augmented Generation, semantic search, vector databases, and backend API development. GitHub Repository: https://lnkd.in/gGsGhwfC #MachineLearning #ArtificialIntelligence #RAG #LLM #FastAPI #Python #DataScience #AIProjects #VectorDatabase

🔬 I built a Production-Grade RAG System from Scratch — Here's How. A research assistant that answers questions with EXACT page citations. How I Built It: 📄 Step 1: Document Ingestion • Load PDFs using PyPDFLoader • Split into semantic chunks (1000 chars, 200 overlap) • Each chunk = 1 searchable unit 🔢 Step 2: Vector Embeddings • Convert text chunks to numerical vectors • Used sentence-transformers (all-MiniLM-L6-v2) • Similar meaning = closer vectors in space 🔍 Step 3: Vector Search • User question → converted to vector • Cosine similarity search across 70+ chunks • Retrieved top-k most relevant chunks 🤖 Step 4: LLM Generation • Retrieved chunks = context • Google Gemini API generates answer • Answer based ONLY on retrieved context • Every answer includes exact page numbers 💾 Step 5: Database & Export • SQLite stores all Q&A pairs • Bookmark important answers • Export to CSV for research documentation Technical Challenges Overcome: Challenge 1: Rate Limits → Implemented retry logic with exponential backoff → Optimized model selection for performance Challenge 2: Slow Startup (25 seconds) → Implemented caching for embeddings → Reduced to 2-3 seconds startup time Challenge 3: Section Detection → Built regex patterns for Roman numerals & numbering → Generated hierarchical tree diagram of document structure Tech Stack: Python | Streamlit | LangChain | Google Gemini API | SQLite | Sentence-Transformers | PyPDF Results: ✅ Processes 12-page papers → 70+ searchable chunks ✅ Sub-3 second response time ✅ 80% faster research analysis ✅ Production-ready web interface #RAG #RetrievalAugmentedGeneration #VectorDatabase #Embeddings #LangChain #GoogleGemini #Streamlit #Python #LLM #GenerativeAI #PortfolioProject

🚀 Project Showcase: Movie Recommendation System I recently built a Movie Recommendation System that suggests movies similar to a user's selected movie using a content-based filtering approach. The system analyzes movie metadata such as genres, keywords, cast, and overview to generate feature vectors using CountVectorizer. It then applies cosine similarity to identify and recommend movies with similar characteristics. To make the project interactive and user-friendly, I deployed it as a web application using Streamlit, where users can select a movie and instantly receive recommended movies along with their posters. The posters are dynamically fetched using the TMDB API. This project helped me gain hands-on experience in: • Machine Learning concepts • Data preprocessing and feature engineering • Recommendation systems • Building interactive dashboards using Streamlit • Deploying data applications to the cloud You can try the live application here: 🔗https://lnkd.in/gdgM-aZX #DataAnalytics #MachineLearning #RecommendationSystem #Python #Streamlit #DataScience #BusinessAnalytics

🚀 Just Deployed My Iris Flower Classification Web Application I recently built and deployed a full-stack Machine Learning application that predicts the species of an Iris flower based on its measurements. What started as a simple ML experiment evolved into building the complete pipeline — from model training to a deployed web application. 🔬 Machine Learning Workflow • Explored and visualized the Iris dataset • Compared multiple models (Logistic Regression, KNN, Decision Tree) • Evaluated models using cross-validation • Selected Logistic Regression as the final model • Saved the trained model using Joblib for inference 🌐 Web Application I integrated the trained model into a Django web application where users can input flower measurements and receive an instant prediction. ⚙️ Engineering & Deployment To make the project production-ready, I also worked on the deployment side: • Built the web backend with Django • Containerized the application using Docker • Used Gunicorn as the production server • Deployed the Dockerized application to the cloud using Render 🔗 Live Application: https://lnkd.in/gdqbuKu3 💻 GitHub Repository: https://lnkd.in/g4dNyEy2 This project helped me understand how Machine Learning models move from notebooks into real-world applications. Tech stack used: Python • Scikit-learn • Pandas • Django • Docker • Gunicorn • Render I’m currently exploring more projects that combine Machine Learning, backend engineering, and deployment. Feedback and suggestions are welcome! #MachineLearning #Python #Django #Docker #AI

I didn’t switch from Flask to FastAPI because of hype. I switched because I got tired of rewriting the same things. At one point, I was building an ML API that needed to handle multiple requests in parallel. What I thought would be a quick setup in Flask turned into: 😮💨 writing validation logic for every endpoint 😮💨 figuring out async handling the hard way 😮💨 plugging in docs separately None of this was “hard”… just repetitive. So I tried FastAPI on the same kind of use case. And the difference was immediate: 🫡 Request validation just worked 🫡 Async didn’t feel like an add-on 🫡 Docs were already there without extra effort It removed a lot of invisible work I didn’t even realise I was doing. Flask still makes sense for: 👍 quick prototypes 👍 smaller apps 👍 when you want full control But for ML APIs that need to scale (or just stay clean as they grow), FastAPI has been a much better default for me. Curious what others are using ? Still on Flask, or moved to FastAPI? Sharing what I learn while building with AI. Follow along 💛 Komal Khetlani #MachineLearning #ArtificialIntelligence #FastAPI #Flask #Python #BackendDevelopment #SoftwareEngineering #Developers

Day 5 Building with AI with Flask⚡ Diving into Flask — a lightweight microframework that keeps things simple and flexible without forcing heavy dependencies. Handling HTTP status codes with a practical example: from flask import request @app.route("/search") def search(): query = request.args.get("q") if not query: return "q is required", 400 if query == "admin": return "Unauthorized access", 401 if query != "flask": return "No results found", 404 return f"Results for {query}", 200 @app.errorhandler(500) def server_error(e): return "Something went wrong", 500 Core dependencies in Flask: Werkzeug — WSGI utilities for handling requests and responses Jinja2 — Template engine for dynamic HTML rendering MarkupSafe — Escapes unsafe characters to prevent injection ItsDangerous — Signs data securely (sessions, tokens) Click — CLI tool support for running and managing apps Popular Flask extensions: Flask-SQLAlchemy — ORM for database interactions Flask-Mail — Sends emails from Flask apps Flask-Admin — Admin dashboard interface Flask-Uploads — Handles file uploads Flask-CORS — Enables cross-origin requests Flask-Migrate — Database migration support Flask-User — User authentication and management Marshmallow — Serialization and validation Celery — Background task processing Flask keeps things minimal while giving complete control over backend architecture. #AI #Flask #Python #WebDevelopment #Backend #LearningJourney #100DaysOfCode#ArtificialIntelligence #MachineLearning #Flask #Python #WebDevelopment #BackendDevelopment #APIDevelopment #FullStack #DeveloperLife #Programming #Coding #SoftwareDevelopment #Tech #TechLearning #LearnInPublic #100DaysOfCode #CodeNewbie #DevCommunity #BuildInPublic #DataScience #AIEngineering #CloudComputing #OpenSource #CodingJourney #Developers #TechCommunity #Innovation #Automation #FutureTech

Landslide Risk Detection & Alert Web App (ML + Flask) Built a web application that predicts landslide risk across 4 classes from user-provided inputs and presents confidence scores visually. The system includes secure authentication and can optionally send real-time SMS alerts when risk is high/moderate. Key features Secure Register/Login with Argon2 password hashing ML-based 4-class risk prediction (predict + predict_proba) Confidence visualization (auto-generated Matplotlib bar chart) Optional Twilio SMS alerts (environment-variable based credentials) Lightweight storage using SQLite Tech Stack: Backend: Python, Flask ML: Pre-trained classification model (pickle / scikit-learn style) Database: SQLite Visualization: Matplotlib Notifications: Twilio SMS Open to feedback and improvements—especially on adding better feature validation, model monitoring, and deployment. #Hashtags #Python #Flask #MachineLearning #DataScience #SQLite #Twilio #Matplotlib #WebDevelopment #AI #DisasterManagement #EarlyWarningSystem #Landslide #RiskAssessment

I gave an AI a website URL. It roasted it.  Here's what I built this week 👇 🌐 WebSnark is an AI-powered website summarizer built with Python, Streamlit, and Google's Gemini API. The idea was simple: → Paste any website URL → Add your free Gemini API key → Get a sharp, witty, markdown-formatted summary in seconds No login. No storage. No fluff. 🛠 Tech Stack: • Python (BeautifulSoup for web scraping) • Google Gemini 2.5 Flash (free tier) • Streamlit (for the live UI) • Custom system prompting for tone control What I learned building this: ✅ Web scraping + LLMs is a powerful combo. You can build a research tool, a competitor analyzer, or a content digest in under 100 lines of Python ✅ System prompts are everything. The same scraped content returns completely different outputs depending on how you instruct the model ✅ Streamlit makes shipping fast. I went from a Jupyter notebook to a live web app in one afternoon ✅ The free-tier Gemini API is surprisingly capable for summarization tasks The whole project started as a Jupyter notebook experiment and turned into a shareable tool anyone can run locally. 🔗 Live demo: https://lnkd.in/gT6nyXUr 💻 Code: https://lnkd.in/gDx4Wyxb If you're learning LLM engineering or want to build your first AI-powered app, start with something small like this. Scrape a page, summarize it, deploy it. That's it. The best way to understand how LLMs work is to build with them. 🚀 #Python #LLM #GenerativeAI #Streamlit #GeminiAPI #AIEngineering #MachineLearning #BuildInPublic #WebScraping #SideProject