AI-Powered Website Generator with Python and FastAPI

🚀 Just shipped a PDF text and image extraction tool. I built a full stack system that converts PDFs into structured outputs you can actually work with. The goal: make it simple to extract both text and visual content from large documents to feed your LLM easily. What it does 📄 Extracts text from PDFs and converts it into clean Markdown (headers, paragraphs, tables) 🖼 Detects and exports figures and tables as separate images 📦 Supports batch uploads with a live progress tracker ⬇️ One-click Download All to export everything as a ZIP Tech stack 🖥 Frontend Next.js 14 (App Router), TypeScript, React, Tailwind — deployed on Vercel 🐍 Backend Python + Flask with a sequential job queue for reliable multi-file processing — deployed on Hugging Face Spaces 🔗 Architecture Next.js API proxy routes backend calls and keeps the HF Space private and secure 📑 PDF processing PyMuPDF4LLM for text extraction + DocLayout-YOLO for layout detection Challenges I ran into 🧩 Tables and figures split across pages → built logic to detect bounding boxes across pages and stitch them into a single image 📝 Pairing images with their captions → added spatial matching between figures and nearby caption blocks ⚙️ Handling multi-file uploads safely → implemented a sequential background queue 🎥You can try a live demo here : https://lnkd.in/dGhQwa6N #DataEngineering #Python #NextJS #PDFProcessing #DataExtraction #FullStackDevelopment #BuildInPublic

Title: Async Views — Handle High-Volume Flower Orders 🚀 Opening Hook: Imagine a bustling flower shop in spring, bouquets flying off the shelves! 🌸 How do you ensure every order blooms with efficiency? The Problem: Traditional views can wilt under pressure. Let's look at the typical approach: ```python # BAD approach def handle\_order\(request\): flowers = Flower.objects.all\(\) bouquets = Bouquet.objects.filter\(order=request.order.id\) # Processing logic return render\(request, 'order.html', \{'flowers': flowers, 'bouquets': bouquets\}\) ``` The Solution: Let your orders blossom using Django's async views! 🌼 ```python # GOOD approach async def handle\_order\(request\): flowers\_task = Flower.objects.all\(\) bouquets\_task = Bouquet.objects.filter\(order=request.order.id\) flowers, bouquets = await asyncio.gather\(flowers\_task, bouquets\_task\) return render\(request, 'order.html', \{'flowers': flowers, 'bouquets': bouquets\}\) ``` Think of it like arranging multiple bouquets at once, instead of one by one! Did You Know? 💡 Under the hood, Django's async views allow for non-blocking I/O operations, which means your server can handle other requests while waiting for database queries. Why Use It? - ⚡ Performance impact - 🧹 Code quality improvement - 📈 Scalability advantage The Golden Rule: Async views will help your code "rose" to the occasion. Engagement Question: How have async views helped your projects? Or what's your favorite tip for using them? Share below 👇 Hashtags: #Django #Python #WebDevelopment #Backend #Performance #FlowerShop #DjangoORM

🚀 Ultima-DL - Built with Vibe Coding I needed a powerful media downloader for personal use, so instead of searching for tools, I decided to vibe code one with AI assistance. The result is Ultima-DL, a lightweight media downloader designed with a clean architecture and fast extraction pipeline. ⚙️ Tech Stack Python + Flask backend Vanilla HTML / CSS / JavaScript frontend yt-dlp for media extraction FFmpeg for merging high-resolution video and audio streams Tkinter for native folder selection What Makes It Interesting • No API required – the system extracts available formats directly from the video page using yt-dlp • Supports Standard, Pro (4K), and Audio-only downloads • Automatically merges adaptive streams using FFmpeg • Fast playlist extraction using lightweight metadata mode • Real-time progress tracking using Server-Sent Events • Temporary download pipeline with automatic cleanup Because the app does not rely on external APIs, it avoids API keys, quotas, and rate limits. Instead, it reads the same streaming data that browsers use to play videos. GitHub Repo: https://lnkd.in/d6_xeAtM #python #vibeCode #OpenSource

The "Personal Progress" Draft 10 days of deep-diving into logic. I’ve been quiet on here for a bit, but the work hasn't stopped. I’ve spent the last week focusing on sharpening my problem-solving logic. Today’s highlight was tackling the Valid Sudoku problem. It’s one of those challenges where the logic is actually quite elegant once you see it. I used a hash map approach with defaultdict(set) and a bit of integer division—(r // 3, c // 3)—to map the 3x3 squares. It’s satisfying when a complex grid starts to feel simple through the right code structure. What’s next? After spending the last few weeks sharpening my JavaScript and DSA logic, I’m officially moving into React today. I've realized that a solid grasp of hashing is the best foundation for state management. Transitioning from 'Vanilla JS' to React components feels like moving from hand-tools to a power-drill. #BuildingInPublic #CodingJourney #Python #JavaScript #WebDevPersonal

🚀 𝗝𝘂𝘀𝘁 𝗘𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗧𝗲𝘅𝘁𝘂𝗮𝗹... 𝗮𝗻𝗱 𝗶𝘁’𝘀 𝗶𝗻𝘁𝗲𝗿𝗲𝘀𝘁𝗶𝗻𝗴 👀 Lately, I’ve been playing around with Textual (Python TUI framework) — no big project yet, just pure experimentation. And honestly… it feels different. 💡 Building UI without a browser 💡 No React, no Angular 💡 Just Python + terminal Still early for me, but a few things stood out: • Super fast to spin up • Clean UI with CSS-like styling • Everything in one language (Python) • Runs anywhere — even over SSH Not saying it replaces web or desktop apps… But for internal tools, dashboards, or admin panels — this could be really useful. For now, I’m just exploring and testing ideas. Let’s see where it goes 𝗔𝗻𝘆𝗼𝗻𝗲 𝗲𝗹𝘀𝗲 𝘁𝗿𝗶𝗲𝗱 𝗧𝗲𝘅𝘁𝘂𝗮𝗹 𝘆𝗲𝘁? #Python #Textual #LearningInPublic #DevExperiment #FullStackDeveloper

I thought FastAPI was fast. Turns out I was just using it wrong. Client called me one day. Why is the API so slow? I thought you said FastAPI is fast? I said yes. FastAPI is async. It handles concurrent requests beautifully. What I didn't say.... My routes weren't async at all. Here's what my code looked like: # What I wrote @app.get("/users/{id}") def get_user(id: int, db: Session = Depends(get_db)): return db.query(User).filter(User.id == id).first() Looks fine right? That def instead of async def — that one word was blocking FastAPI's entire event loop. Every request was waiting for the previous one to finish. Not concurrent. Not async. Just a very expensive queue. The fix was literally one word: # What it should have been @app.get("/users/{id}") async def get_user(id: int, db: AsyncSession = Depends(get_db)): result = await db.execute(select(User).where(User.id == id)) return result.scalars().first() def → async def Response time under load: 4 seconds → 200ms. The lesson nobody tells you clearly: FastAPI is only as async as you make it. The framework won't save you from blocking code. It just gives you the tools — you still have to use them correctly. → Always use async def for route handlers → Use httpx not requests inside async endpoints → Use AsyncSession not Session for DB calls One word. One month of slow APIs. One very unhappy client. #FastAPI #Python #BackendDevelopment #AsyncProgramming #SoftwareEngineering #LessonsLearned #WebDevelopment

Built something that's been on my list for a while 👀 I built an open-source CLI tool called systemgen that lets you  save your entire project stack as a reusable profile and turn it into a  structured prompt any AI coding tool understands immediately. No need to constantly re-explain your setup EVERY single session. How it works: → Run systemgen init and answer a few questions about your stack → Your framework, folder structure, naming conventions + your prompt → Pipe it straight to Claude Terminal, Ollama, or copy to clipboard for Cursor Works for any language and stack: → Python / FastAPI → TypeScript / Next.js → Dart / Flutter → Go, Rust, JavaScript + more Useful when you want your AI to already know how you work before  you write a single line :) Pair it with mem0 = no starting from scratch, no re-explaining decisions you already made 🧠 w/ https://lnkd.in/eV7j5ScK Install 🔗: pip install systemgen Source 🔗: https://lnkd.in/e9FcDhdN Hope this helps ✅ Drop a Like if you found this useful! 👍 Follow me for more projects like this in the future💎 #python #opensource #ai #developertools

Currently building a Geographic Information System(GIS)-based application and it has pushed me slightly outside my usual stack. I mostly work with JavaScript (React, modern frontend tools), but for this project I need OpenCV, which works better with Python. So I’m learning Flask to integrate the computer vision component with the web app. Interesting challenges so far: . Structuring communication between Python backend and JS frontend • Managing performance when processing images • Keeping the architecture simple while mixing technologies One thing I’m learning again is that real-world projects often require flexibility beyond your primary stack. If anyone has experience combining Flask + frontend frameworks, I’d appreciate any tips or best practices.

🚀 Starting my journey in web development! I’m happy to share my first project — a Flask CRUD Web Application 💻 🔹 Built using: Flask, MySQL, HTML, CSS 🔹 Features: Create, Read, Update, Delete operations 🔹 Deployed using Render 🔹 Source code available on GitHub 🌐 Live Demo: https://lnkd.in/gXdhjUYY 💻 GitHub Repo: https://lnkd.in/gRbe_2YB This is just the beginning — looking forward to building more advanced and impactful projects ahead! #Flask #Python #WebDevelopment #BeginnerProject #svhec Dr. Muralisankar Kumaresan Guide KABILESH RAMAR

Today, I’ve been working on Django Views to bridge the gap between user input and the database. 💻 Key Learning: Why use commit=False? I learned that using item = form.save(commit=False) is crucial when you need to modify an object before saving it to the database. In this case, I used it to manually attach the currently logged-in user to the new item. Other implementations today: ✅ Access Control: Secured views with the @login_required decorator. ✅ Dynamic Queries: Used get_object_or_404 for better error handling and filtered related items for a better UX. ✅ Request Handling: Managed the flow between GET (viewing the form) and POST (submitting data). Backend development is like solving a puzzle—every piece of logic matters! 🧩 #Django #Python #BackendDeveloper #CleanCode #SoftwareEngineering #CodingJourney #WebDevelopment