python
Shipped: AI-Powered Product Review Analyzer
I built an AI-powered Product Review Analyzer that turns customer reviews into useful insights. The goal was to make it easier to understand what customers actually think about a product without manually going through hundreds of reviews. What it does Analyzes product reviews Identifies customer sentiment Extracts useful insights from review data Presents the results through a simple interface What I learned 🤖 Working with AI and NLP 📊 Processing real-world review data 💻 Building an end-to-end AI application 🎨 Improving UI/UX for an AI tool 🧩 Turning an idea into a working product I'm still improving the project and would love feedback from other builders. GitHub: https://github.com/nidhishreebai-gif/product-review-analyzer What feature would you add to make this more useful? 🚀
Fake Review Detection System
Fake Review Detection is a machine learning-based system designed to detect fake and genuine online reviews. Using Python and machine learning techniques, the system analyzes review text and identifies patterns that may indicate fraudulent or misleading reviews. The goal is to help users and businesses distinguish authentic feedback from manipulated content and make more trustworthy decisions.
OllamaBench
A local LLM benchmarking platform for evaluating and comparing models running through Ollama. It automatically measures: Performance and response latency Quality across reasoning, coding, creative writing, and factual tasks CPU, GPU, and memory usage Multi-model performance comparisons Exportable benchmark reports Everything runs locally with no cloud dependency, keeping benchmark data private. Built with React, TypeScript, FastAPI, Python, SQLite, and Ollama. I built this to answer a practical question: Which local LLM performs best for a specific workload and hardware setup? GitHub: https://github.com/Fortuner47/OllamaBench Would love feedback on the benchmarking methodology and what metrics or evaluation methods I should add next.
GestureFX – AI Hand Gesture Visual Effects
Building GestureFX – an AI-powered hand gesture recognition system using Python, OpenCV, and MediaPipe. The project detects hand gestures through a webcam and triggers real-time superhero-inspired visual effects such as: - Energy Shield - Energy Beam - Energy Blast - Portal Effect Exploring Computer Vision, Human-Computer Interaction, and real-time gesture tracking to create a touchless interactive experience.
Built an AI-Powered Meeting Intelligence Platform - ContextBridge
We've all been in the meeting after the meeting: "Wait, what did we actually decide? Who owns that? Didn't someone raise a risk we never resolved?" ContextBridge reads a raw transcript and automatically pulls out: ◆ Decisions — with owner and a confidence score ○ Action items — with assignee, blockers, and deadlines △ Gaps — the risks nobody flagged out loud (an unowned deadline, an unresolved disagreement) Then it lets you ask questions in plain English across every meeting — like "what's blocking the launch?" and streams the answer back, citing which meetings it came from. Check it out here -- [https://contextbridge.dhruvdeshpande.in/
🚀 Built a House Price Prediction Model using Python & Machine Learning
I built a House Price Prediction project as part of my Data Science Internship at Oasis Infobyte . 📌 Project Overview The goal of this project is to predict house prices based on different property-related features using Machine Learning . 🔄 Project Workflow Dataset → Data Cleaning → EDA → Feature Selection → Model Training → Evaluation → Price Prediction 🧹 Data Preparation Loaded and explored the dataset using Pandas Handled missing and inconsistent data Prepared the dataset for machine learning 📊 Exploratory Data Analysis Analyzed relationships between different features Created visualizations to identify patterns and trends Studied factors affecting house prices 🤖 Machine Learning Selected relevant features Split the dataset into training and testing sets Trained a regression model Evaluated the model using appropriate performance metrics 🎯 Prediction The trained model takes property-related features as input and generates an estimated house price . 🛠️ Tech Stack Python Pandas NumPy Matplotlib Scikit-learn Jupyter Notebook 💡 Key Learning This project gave me hands-on experience with the complete machine learning workflow — from data preprocessing and visualization to model training, evaluation, and prediction . 🔗 GitHub Repository 👉https://github.com/pallavisagar07/OIBSIP/tree/ec73e94f0a9d2ea3a3fe3e65c1676d8c0af17219/DataAnalytics-L2-House-Price-Prediction-Linear-Regression Python MachineLearning DataScience DataAnalytics ScikitLearn OasisInfobyte ProjectShowcase
Level 1 AI Agent in n8n (Gemini + Tools + Memory)
Just built a custom Level 1 AI Agent using n8n! 🚀 Here is what's running under the hood: - LLM Model: Google Gemini Chat Model - Memory: Window Buffer Memory for conversation context - Tools: Custom HTTP Request tool fetching live web data (Wikipedia API) & Calculator tool In this demo, the agent autonomously decides to call the HTTP tool to fetch live information about Tesla and formats the output smoothly. Feedback and suggestions for Level 2 features are welcome!
Hey DevConnect! Building AI Voice Agents & Workflow Automations
Hey everyone! I'm Ali Hassan , a Computer Engineering student at UET Lahore focused on building production-ready AI systems and agentic workflows . I specialize in bridging full-stack web development with deep automation pipelines—moving past basic demos to build systems that handle real operations end-to-end. 🛠️ The Stack Behind My Automations: • AI Systems & Voice: Vapi, ElevenLabs, Gemini, Claude, OpenAI, n8n, Pinecone, Lovable AI, v0, Bolt.new • Languages & Web: Python, C++, SQL, JavaScript, TypeScript, React, Node.js, FastAPI, Tailwind CSS, REST APIs • Integrations & Cloud: Green API, Twilio, Airtable, Docker, Vercel, Database Design, Google Sheets, Notion API 💡 What I'm currently building: 1. Production AI Voice & Support Agents: Conversational pipelines with sub-second latency, automated CRM sync, and dynamic lead scoring. 2. Academic AI App: An intelligent platform leveraging LLMs to optimize study workflows and context-aware Q&A for students. Excited to connect with fellow AI builders, showcase technical builds, and collaborate! 📁 Portfolio: https://www.alihassan-builds.me/
CipherVault — Multi-Technique Encryption & Decryption Lab
I built CipherVault, an interactive Python/Flask cryptography application that brings multiple encryption and decryption techniques into one cybersecurity-focused workspace. It supports Caesar, Vigenère, Atbash, Rail Fence, XOR, Fernet, and AES-256-GCM, along with a learning section, live transformation visualizations, side-by-side cipher comparison, encryption statistics, and password-strength feedback. The goal was to go beyond a basic encrypt/decrypt tool and create something that is also useful for understanding how different cryptographic techniques work and how their security characteristics differ. I'd love to get feedback from other developers and security enthusiasts on the project and ideas for where I could take it next. cybersecurity cryptography python flask infosec
CodeMorph
I built CodeMorph , a Python project that explores custom text encoding and decoding through a simple command-line interface. The project combines text transformation with audio-based interaction, allowing encoded data to be worked with through a lightweight CLI workflow. What I explored: 1. Custom encoding and decoding logic 2. Text transformation 3. Morse Style communication concepts 4. Audio input/output 5. Command-line interaction 6. Python project structure Tech Stack: Python · NumPy · SoundDevice · Colorama The main goal of CodeMorph was not just to build another utility, but to understand how different components can come together to create a small, functional communication tool. GitHub: CodeMorph I’m sharing it here to document the project and get feedback on how the implementation could be improved or extended.
Shipped: Nexora - Intelligent AI Web Assistant
Built Nexora, an intelligent AI web assistant designed for seamless, intuitive, and efficient digital interaction. Key Features: Contextual AI capabilities with integrated REST APIs for real-time processing Clean, minimalist UI optimized for smooth human-AI interaction Web search grounding and voice input support Modular Python architecture deployed live via Vercel Tech Stack: Python, REST APIs, Vercel, GitHub Would love to hear your feedback and suggestions from the community!
TermTint
Coloring terminal output sounds simple… Until you actually try to build it. Making terminal colors work programmatically can get surprisingly complex, especially when you want the output to stay clean, consistent, and readable. So, I started building something around that idea: TermTint A Project focused on bringing color, clarity, and a better visual experience to the terminal. It’s still under development. Still experimenting. Still breaking things. Still fixing things. Still making it better. But the results are starting to look interesting. This is just a sneak peek. Something colorful is coming to the terminal.
TermTint
Just shipped TermTint — a lightweight, zero-dependency Python library for colored and styled terminal output. No bloated dependencies, no config headaches — just clean, readable terminal styling that works out of the box. ✅ 9/9 CI checks passing ✅ Tested across Python 3.9 → 3.14 ✅ Verified on Ubuntu & Windows ✅ MIT licensed and open source Building small, focused tools like this is one of my favorite ways to learn, solving a real annoyance (unreadable terminal output) with something simple enough to actually maintain.
🚀 Built a Basic RAG and an Advanced RAG System from Scratch — Without LangChain
To understand how RAG works internally, I first built a Basic RAG Pipeline: Basic RAG PDF → Extraction → Chunking → Embeddings → FAISS Search → Top-K Chunks → Prompt → LLM → Answer It works well for simple questions, but struggles with exact keywords, IDs, and complex queries. So I built an Advanced RAG Pipeline: Advanced RAG PDF → Extraction → Chunking → Embeddings → Dense + Sparse Index → Query Understanding → Dense Retrieval + Sparse Retrieval → Reciprocal Rank Fusion (RRF) → Cross-Encoder Reranking → Noise Filtering → Context Fusion → Prompt → LLM → Answer Key Improvements ✅ Semantic Search (FAISS) ✅ Exact Keyword Search (BM25) ✅ Hybrid Retrieval ✅ Cross-Encoder Reranking ✅ Duplicate Removal ✅ Better Context Selection ✅ Reduced Hallucinations One of the biggest lessons from this project: A better RAG system isn't always about using a better LLM. It's often about building a better retrieval pipeline. Tech Stack: Python • Streamlit • FAISS • Sentence Transformers • BM25 • Cross-Encoder Reranker • Groq RAG GenAI LLM AIEngineering MachineLearning Python FAISS BuildInPublic
Shipped: An AI Job Matcher Agent — paste a job URL + your resume, get a match score, skill gaps, and a cover letter
Applying for jobs is the same loop every time: read the posting, compare it with your resume, guess your chances, then write a cover letter from scratch. I built an agent that does all of it in one request. Flow: paste a job posting URL → upload your resume PDF → get the analysis back. Output: - match score (0-100) — how well the profile fits the role - skill gaps — what the job asks for that the resume doesn't have - strengths — what actually lines up - cover letter — personalized, ready to send Stack: - FastAPI — async API, two endpoints: /analyze (JSON, local resume path) and /analyze-upload (multipart file upload). Auto Swagger docs at /docs . - OpenAI GPT — the reasoning layer. Single structured prompt, JSON-only output, validated with Pydantic before it ever reaches the response. - httpx + BeautifulSoup — scrapes the job page, pulls the title and hunts for requirement lists (looks for requirements / responsibilities / you will headings, falls back to the first bullet lists on the page). - pdfplumber — page-by-page text extraction from the resume PDF. - Jinja2 + vanilla CSS — drag-and-drop UI, no frontend framework. - Docker — python:3.12-slim , single container, runs anywhere. Two things I'd call out: 1. Graceful degradation. If there's no API key or the LLM call fails, it doesn't 500 — it falls back to a keyword-based scorer that intersects job skills with resume skills and returns a real score. The app is always usable. 2. Scraping is the hard part, not the AI. Every job board renders differently, and half of them are JS-heavy. The heuristic parser works on plain HTML postings but site-specific parsers (LinkedIn, Indeed) are the obvious next step. Next up: semantic matching with embeddings instead of keyword overlap, batch mode for multiple postings at once, and swapping temp-file handling for in-memory PDF parsing. Happy to answer anything about the scraping heuristics or the fallback design. I mostly build AI agents and automation for businesses — chatbots, workflow automation, the boring repetitive stuff that shouldn't be manual. If you're working on something similar, DMs are open. Portfolio: https://danish614.github.io/danish-portfolio/
Working on a Sentinel-2 Ground Station Project
Currently stucked while working on an experimental project to better understand how Sentinel-2 satellite data can be received, processed, and analyzed. As part of this project, I’m trying to design and build a suitable antenna and integrate it with an SDR setup, laptop, and Python-based software. My goal is to explore satellite communication, RF signal analysis, demodulation, satellite tracking, and Earth-observation data processing. The project will include: - Antenna design and signal reception. - Monitoring RF signals using SDR hardware. - Python-based signal processing and automation. - Satellite pass tracking and Doppler-shift analysis. - Decoding, visualizing, and interpreting authorized or publicly available data. - Processing Sentinel-2 imagery using Python tools and libraries. Since Sentinel-2’s operational payload downlink uses high-data-rate X-band communications, directly receiving the satellite’s raw downlink requires specialized equipment and must comply with local regulations. Therefore, I’m initially focusing on publicly available Sentinel-2 data, recorded signals, and simulation-based experiments. My objective is not to interfere with any communication system, but to learn more about satellite communications, remote sensing, RF engineering, and Python-based data analysis. Sentinel-2 data is openly available through the Copernicus Data Space Ecosystem and can be incorporated into Python workflows. If you have experience with antenna design, SDR, satellite tracking, RF signal processing, or Python-based Earth-observation analysis, I would appreciate your suggestions and guidance. Sentinel2 SatelliteCommunication SDR Python RemoteSensing EarthObservation AntennaDesign SignalProcessing Copernicus SpaceTechnology
TermTint v0.3.0
Terminal output shouldn't need a heavy framework just to look good. I'm excited to share TermTint v0.3.0 - a lightweight, zero-dependency Python library for terminal colors and styling. With v0.3.0, TermTint now includes: • Reusable styled() API • Custom themes • RGB / True Color support • 256-color ANSI support • Multiple style combinations • Windows + Unix terminal support • NO COLOR / FORCE COLOR support • 89 tests with 100% coverage • Python 3.9+ support The goal is simple: Keep terminal styling powerful enough to be useful, while keeping the library small, predictable, and dependency-free. This release also keeps the existing APIs from previous versions intact, so hashtag TermTint can grow without breaking the simplicity that it started with. I'm building hashtag TermTint as an open-source project and continuing to improve it release by release. GitHub: https://github.com/hasheramin5-cyber/TermTint PyPI: https://pypi.org/project/termtint/ v0.3.0 is out. What's one terminal feature you think a lightweight styling library should have next? Python OpenSource SoftwareDevelopment GitHub DeveloperTools Programming
HeritageAI
Building HeritageAI 🎙️🤖 — a voice-controlled PowerPoint assistant that lets presenters navigate slides hands-free using natural voice commands. Currently improving it with smarter speech interpretation and AI-powered presentation features. 🚀 🔗 https://bit.ly/4Aj9LP0 AI MachineLearning VoiceAI Python HeritageAI
Shipped: E-Commerce Price Tracker watches Amazon & Daraz, emails you when the price drops
Ever kept refreshing a product page waiting for a sale? I automated that. Paste any Amazon / Daraz product URL + your target price, and it scrapes the live price, stores the history, re-checks every few hours, and emails you the moment it hits your target. What it does - 🔎 Live price scraping (works behind bot protection) - 📉 Full price history per product - ⏰ Auto re-check every few hours (Celery Beat) - 📧 Email alert when target price is hit - 🌐 REST API + Web UI + CLI Tech stack - Backend: Python, FastAPI, Pydantic - Background jobs: Celery + Celery Beat + Redis - DB: PostgreSQL, SQLAlchemy, Alembic migrations - Scraping: Selenium, undetected-chromedriver, BeautifulSoup, cloudscraper - Alerts: SMTP + Jinja2 templates Hardest parts - Anti-bot protection — plain requests gets blocked instantly. Ended up with undetected-chromedriver in headless mode, cloudscraper as a lighter fallback. - HTML keeps changing — Daraz especially. Built multi-selector fallback logic so one broken selector doesn't kill the whole scrape. - Parallel checks — moved from a loop to a Celery task queue so hundreds of products get checked concurrently. - Proper schema — versioned Alembic migrations, not a throwaway script. Same architecture works for competitor price monitoring, stock/availability alerts, or any "watch a site and notify me" automation. 🔗 Code: https://github.com/danish614 🌐 Portfolio: https://lnkd.in/dXE58DgU
One repository. Hundreds of programs. One long Python learning journey.
One repository. Hundreds of programs. One long Python learning journey. I’ve been building something on GitHub that started as a Python practice repository, and it has grown into much more than that. Python-Practice-Repository is my structured learning repository where I’ve been systematically working through Python from fundamentals to advanced concepts. So far, I’ve completed 17 major sections, covering areas such as: • Python Basics • Conditions & Loops • Functions • Strings, Lists, Tuples, Dictionaries & Sets • File Handling • Exception Handling • OOP • Modules • Regular Expressions • Iterators & Generators • Decorators • Multithreading Each section contains structured programs designed to build understanding through practice, not just theory. And now, there’s only one final phase left: 18. Projects/ This is where the learning turns into building. The goal is to create real-life, practical Python projects that combine the concepts I’ve learned throughout the repository and solve problems that actually make sense outside a classroom. But I don’t consider the repository finished after that. Going forward, I’ll continue evolving it with useful notes, resources, assets, references, and additional learning material as I discover and build them. What started as practice is gradually becoming a long-term Python learning resource, both for myself and hopefully for anyone else who finds it useful. 17 sections completed. 1 final phase remaining. The real building starts now. GitHub: https://github.com/hasheramin5-cyber/Python-Practice-Repository Python GitHub LearningInPublic OpenSource SoftwareDevelopment Programming DeveloperJourney ContinuousLearning
Personal Task App: DevPort
I built a control center for all the dev servers on my Windows machine. I keep 40+ repos in one GitHub folder, and I kept losing time to the same questions: which terminal is running what, which port is taken, why did the API die, and what was I supposed to work on next. DevPort puts all of that in one window. What it does - Finds my projects on its own. I pick a folder and it scans for repos, then works out what can run: Vite and Next apps, plain Node servers, FastAPI backends, PowerShell dev scripts. It only reads config files. It never runs repo code to figure things out. - Starts and stops services in order. The backend comes up before the frontend, and it waits for each one to be ready before moving on. - Handles port conflicts. If a port is taken, it shows which process owns it and suggests a free one instead of failing. - Checks health. It uses HTTP or TCP checks, and three failures mark a service as degraded. Optional auto-restart backs off (2s, 5s, 15s), then stops trying. - Keeps logs. You can search, filter and download them. Secrets are redacted before anything is written to disk. - Has a to-do planner built in. It supports subtasks, dependencies, dates and tags, plus a Gantt view that draws the dependency links. It rejects circular dependencies.
Air Rhythm — Play music with your hands
Happy to share one of the projects I have been working on: Air Rhythm. It’s a computer vision rhythm game where you play music by moving your hands in front of a webcam. In this demo, I’m playing a short version of Für Elise by catching falling notes with my fingertips. How it works: • Your fingertips control virtual drumsticks. • Catching a falling circle plays its musical note. • Following the note sequence lets you build the melody yourself. • A live camera inset shows your hands and the tracked skeleton behind the interaction. The computer vision behind it: I used MediaPipe’s pretrained Hand Landmarker to detect 21 landmarks on each hand, then built the interaction around those predictions: • Adaptive smoothing to reduce tracking jitter. • Collision checks along fingertip movement between camera frames. • Time-based note movement and sequence-aware scoring. • Synthesized audio triggered by successful hits. Tech stack: • Desktop: Python, OpenCV, MediaPipe Hand Landmarker, NumPy, sounddevice, and pywebview. • Browser: TypeScript, OpenCV.js, MediaPipe Hand Landmarker, and Web Audio. • Shared interface: HTML, CSS, and Canvas. • Hosting: GitHub Pages. Most of the testing went into making quick movements register and reducing jitter. Partially visible hands still challenge the tracker, so the interface prompts users to keep their hands and wrists in view. Both versions process camera footage locally on your device without uploading it. Try the live web version: https://adolph1999v.github.io/air-rhythm/ GitHub repo, code, and documentation: https://github.com/Adolph1999v/air-rhythm Give it a play and let me know what you think 😄