I build web applications and AI systems that actually work in production.
My focus lately is RAG pipelines, LLM fine-tuning, and making AI useful for people who don't have access to expensive legal or technical help.
| Domain | Technologies |
|---|---|
| AI & ML |    RAG, Vector Search (FAISS, Pinecone), LLM Integration |
| Frontend |       |
| Backend |    |
| Database |   |
| DevOps |    |
| Project | Description | Stack |
|---|---|---|
| twitter-automation-ai | Python framework for Twitter/X automation — multi-account scraping, posting, engagement, and content analysis using Selenium and LLMs (OpenAI, Gemini). | |
| wpaudit | WordPress security auditing suite. Automates pentesting with Nmap, WPScan, Nuclei, and SQLMap. Outputs full reports. Built for ethical hackers. | |
| llmoptimizer | Generates llms.txt files for websites and docs. Supports crawling, sitemaps, static builds, and adapters for Next.js, Vite, Nuxt, and Astro. |
 |
| seo-analyzer | On-page, technical, and content SEO audits via CLI or Flask API. | |
| student-performance-analysis | Data science project analyzing student performance using XGBoost, SHAP, and K-Means clustering for segmentation. |  |
| yolo-image-identifier | Real-time object detection using YOLOv8, FastAPI, and React. High-contrast visualization, built for industrial use. |  |
| ml-mastery-path | Reference implementations covering deep learning, PyTorch, and generative AI. Built as I learned — still adding to it. |  |
- Lexiber AI — Free legal AI for Pakistan. Lets citizens ask legal questions in English or Urdu and get actual answers, not a referral to a lawyer they can't afford. Built under the UN SDG Access to Justice framework.
"Code is like humor. When you have to explain it, it's bad." – Cory House