AI-Powered Website Summarizer Built with Python and Gemini API

I recently developed a web application that can extract content from any website and generate a clean, structured summary using AI. 🔍 Key Features: • Smart web scraping (removes ads, scripts, clutter) • AI-powered summarization • Clean and readable output • Simple and interactive UI 🛠 Tech Stack: Python | BeautifulSoup | Streamlit | OpenAI API This project helped me combine web scraping with AI to build something practical and useful. 🔗 GitHub: https://lnkd.in/dYY5BvrV 🌐 Live Demo: https://lnkd.in/d4WM_Gwj #Python #AI #WebScraping #MachineLearning #Streamlit #OpenAI #Developers #Projects

Built an AI-powered web crawler using Python, Crawl4AI, and Groq The system scrapes multi-page marketplace listings and converts unstructured web content into structured data such as name, location, price, capacity, rating, reviews, and descriptions. Key things I worked through while building this: 1. asynchronous web scraping with pagination 2. Integrating LLMs for structured extraction 3. Handling blocked page navigation 4. Managing API rate limits and model deprecation 5. Exporting clean datasets for analysis Tech stack: Python, Asyncio, Crawl4AI, Groq, LiteLLM This project gave me hands-on experience combining web scraping with AI for real-world data extraction. Compared to traditional scraping approaches that require extensive manual parsing and brittle selectors, this solution uses LLM-based extraction to convert unstructured content into structured data more flexibly. GitHub: [https://lnkd.in/gMZwN_Ei]

I built an AI agent from scratch. No LangChain. No LangGraph. No CrewAI. Just Python, Gemini 2.5 Flash, and raw tool calling. Here's what I learned that no framework tutorial will teach you: 1. The agentic loop is embarrassingly simple Build messages → call LLM → if tool_call → execute → feed result back → repeat. That's it. Every framework is just a wrapper around this. Once you see it raw, you can never unsee it. 2. Frameworks hide your bugs from you When something breaks in LangChain, you're debugging the framework. When something breaks in raw Python, you're debugging your logic. Big difference. One makes you smarter. One makes you dependent. 3. Tool schema design is where agents actually fail The LLM doesn't call the wrong tool because it's dumb. It calls the wrong tool because your schema description was ambiguous. Write your tool descriptions like you're explaining them to a junior dev on their first day. Precise. No assumptions. 4. 50 lines of Python is enough to go to production My personal concierge agent — the one that lives on my portfolio, captures leads, and pings my phone instantly — is ~50 lines. No overhead. No magic. Just code I fully understand and can debug at 2am. 5. You should build one without a framework at least once Not because frameworks are bad. LangGraph is excellent. I'm using it next. But if you've never written the raw loop yourself, you're flying blind. You're trusting abstractions you don't understand. Build it raw first. Then use the framework. You'll use it 10x better. --- Full source code in the comments — ~50 lines, no magic, just the loop. Follow along if you're into agentic AI and building real things, not just demos. #AgenticAI #Python #BuildingInPublic #LLM #SoftwareEngineering

Everyone asks which Python backend to learn. I've used all three in production. Here's the answer no one gives you. 🔷 FLASK Pros: Total freedom, minimal, very learnable Best for: Learning, tiny internal tools Honest truth: You'll outgrow it fast. And that's fine. 🐍 DJANGO Pros: Full-featured, brilliant ORM, auth built in Best for: Full web products, SaaS, content platforms Honest truth: Heavyweight for APIs. Brilliant for what it's designed for. ⚡ FASTAPI Pros: Async native, blazing fast, auto-generates docs, Pydantic validation Best for: AI/ML APIs, microservices, anything real-time Honest truth: If you're building AI systems in 2026 and you're not using FastAPI, you're making your life harder. My verdict: → AI/ML/microservices: FastAPI. Every time. No debate. → Full web product: Django. → Just learning? Start Flask for one project so you understand the fundamentals. Then move to FastAPI and never look back. Stop debating. Pick one. Build something this week. Which are you currently using — and what are you building? Drop it below. I'll tell you if you've made the right call. #python #fastapi #django #flask #webdevelopment

6 data sources. 1 python script. $0 per month. I built an insights engine that pulls from PostHog, dub.co, Waitlister, Beehiiv, and - cross-references everything - and tells me exactly which channel drives signups. here's why this matters more than most founders realize: for 3 weeks I was running TikTok ads, posting on X, sending DMs, scheduling content - and had zero idea which one actually converted. all channels looked "active." none had proper attribution. then I set up ?ref= tags on every single link. every channel, every bio, every PostHog tracks the full funnel: visit to CTA click to signup. first insight after turning it on: 170 visits from X had no ref tag because the bio link was untagged. I was blind to my biggest channel. second insight: relatively high bounce rate. 56% leave within 5 seconds. third insight: 15% visitor-to-signup conversion for people who actually stay. meaning the product message works - the first impression doesn't. attribution is so important at this stage, so you can iterate fasrt

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

I've been utilising AI tools like Claude to assist with writing Python code recently, and I've encountered some subtle issues. In a single scan of a FastAPI app built with AI assistants, I identified several concerns: - A blocking open() inside an async def that silently froze the event loop - Circular imports surviving only by luck of Python's resolution order - A web-hook signature bug that passed every test because the tests had the same bug - A 78-complexity god function containing the entire API as nested closures None of these issues caused crashes, yet all were shipped to production. AI writes confident, working code, but it’s not always code you'd want to debug at 2 AM. In my quest to identify these specific code smells—beyond just PEP 8 violations—I discovered react-doctor for React projects and questioned why Python lacked a similar tool. This prompted me to create one. The development process took about a day (well, half a day and then a few more hours for "just one more rule"). Interestingly, I used Claude to help develop it, leveraging AI to create a tool that checks AI-generated code. The highlight? I ran it on its own source code and achieved a score of 86/100. I refined my tool before shipping it, ultimately getting it to 100. Initially, I wanted to name it python-doctor, but that name was already taken on PyPI. I compared PyCodeGate and python-doctor: python-doctor aggregates existing tools (Bandit, Ruff, Radon) into a health score, while PyCodeGate is specifically designed for AI-generated code with framework-aware rules, nine dedicated security detectors for AI-prone pitfalls, a diminishing-returns scoring system to prevent one noisy category from negatively impacting your score, SARIF output, dead-code analysis, and native integration with five AI coding agents. I ultimately chose pycodegate, which I now prefer. Open to pull requests and discussions. Share it if you find it useful. It's live and ready for use. Simply point https://lnkd.in/g8FjpBeF pip install pycodegate #Python #OpenSource #AItools #DeveloperTools #SoftwareEngineering #StaticAnalysis #Anthropic #AI #PythonDevelopment

I built a full-stack AI appointment booking assistant — with ₹0 in API costs. No OpenAI. No monthly subscriptions. Just Python, FastAPI, and Ollama running a 120B parameter model. Here's what it does: → Chats with patients naturally via a browser UI → Recommends dental services based on symptoms → Checks real-time slot availability from a live database → Books appointments and stores them permanently → Prevents double-booking automatically The full stack: • Backend → Python 3.11 + FastAPI • Database → SQLite + SQLAlchemy ORM • AI Model → Ollama gpt-oss:120b-cloud • Frontend → HTML + CSS + Vanilla JS • Server → Uvicorn I wrote a complete step-by-step tutorial covering: ✅ Ollama installation & model setup ✅ Python virtual environment setup ✅ Full database schema design & seeding ✅ All 6 REST API endpoints ✅ The complete chat UI from scratch ✅ GitHub deployment No prior AI experience needed. If you know basic Python, you can follow along. 📖 Full article on Medium → https://lnkd.in/dNGnGykT 💻 Complete source code on GitHub → https://lnkd.in/dKp8JkYH If you're learning Python backend development or want to build AI-powered apps without cloud bills, this is a good starting point. #Python #FastAPI #AI #Ollama #OpenSource #WebDevelopment #SQLite #LLM #BackendDevelopment #Tutorial #Beginner #AiAssitant #AiPoweredApp

🚨 Project Launch: Fake News Headline Generator Ever wondered how easily headlines can be created—and sometimes manipulated? I built a fun yet thought-provoking project using Python + Streamlit that generates random “breaking news” headlines in real time. While it looks simple, it highlights an important concept: 👉 Not everything that looks like news is real. 🔍 What this project demonstrates: • Dynamic content generation using Python • Interactive dashboard building with Streamlit • UI/UX customization with CSS • Concept behind automated media systems 💡 Why this matters: In today’s data-driven world, automation can create content at scale. This project is a small step toward understanding how fake news, clickbait, and automated media can be generated—and why data awareness is critical. ⚙️ Tech Stack: Python | Streamlit | Random Module 🚀 This is part of my journey into Data Science & real-world applications of automation. I’ll be improving this by adding: • NLP-based headline generation • Sentiment analysis • Real-world data integration 🔗 Would love your feedback and suggestions! #DataScience #Python #Streamlit #Projects #MachineLearning #AI #LinkedInLearning #BuildInPublic

From Simple Script to Real Learning — My Web Scraping Journey I recently worked on a Python-based web scraping data, and what started as a simple task quickly turned into a powerful learning experience. While extracting data, I faced several challenges: • Handling dynamic web content • Dealing with inconsistent HTML structures • Ensuring the script runs reliably across multiple executions Instead of giving up, I kept iterating, debugging, and improving my approach. Each version of my script became more accurate, efficient, and stable. Tools & Technologies Used: • Python • BeautifulSoup • Requests • Debugging and iteration techniques This project helped me understand how real-world websites behave and how to adapt scraping logic accordingly. Key takeaway: Real learning happens when things don’t work the first time. Looking forward to building more such practical projects. #WebScraping #PythonProjects #DataExtraction #LearningByDoing #TechJourney