r
ratnesh2507

Ratnesh Shukla

@ratnesh2507

Full Stack Developer AI and Data Solutions

India
English, Hindi
About me
I'm Ratnesh Kumar Shukla, a Full Stack Developer specializing in modern websites, landing pages, AI solutions, and data automation. I build fast, responsive, and visually polished web experiences using React, TypeScript, Node.js, Python, and Tailwind CSS. I also develop custom APIs, web scraping, data and media extraction solutions for static and dynamic websites, automation tools, OCR workflows, and backend services. Portfolio: https://ratnesh-shukla.vercel.app. Feel free to message me to discuss your project requirements.... Read more

Skills

r
ratnesh2507
Ratnesh Shukla
Offline • 
Average response time: 2 hours

See my services

Data Scraping
I will do web scraping and data extraction
From$30 / projectMore details
Custom Websites
I will design a modern portfolio website
From$200 / projectMore details

Portfolio

Work experience

Self_Employed

Freelance Backend & Data Automation Engineer

Self Employed • Self-employed

Dec 2024 - Present • 1 yr 7 mos

Architected and deployed an end-to-end media ingestion pipeline to scrape, process, and migrate 2,000+ episodes, successfully managing the high-speed transfer of over 5TB of video data to Google Drive using rclone optimization. Engineered an advanced web scraper using Node.js and Puppeteer, connecting to remote browser instances to bypass geo-restrictions and intercept internal XHR network requests for secure HLS (.m3u8) stream extraction. Built a high-throughput, multi-threaded Python backend using Flask and ThreadPoolExecutor to concurrently download and stitch 4MB video chunks, optimizing bandwidth utilization for continuous 150 Mbps network execution. Designed a fault-tolerant, resumable architecture utilizing SSD-to-HDD caching, automated directory polling, and file-based state tracking (.done markers) to ensure zero data loss during long-running execution cycles. Automated complex data mapping workflows by integrating dynamic Node.js scripts with Excel-based localized metadata, parsing thousands of URLs and mapping them directly to the download queue. Managed large-scale media processing workloads (multi-GB to multi-TB scale) across multiple execution cycles with a reliability-focused execution design.