Mohammad Fraz

DSA is just the beginning! 👀🔥 From DSA to System Design, get ready for the next level of interview preparation. INTERVIEW 360 - DSA to System Design [DSA, LLD, HLD, Machine Coding, System Design, SDE Interviews, Placements]

2026 graduate cracked 25LPA offer

CS fundamentals

Coding interview preparation sheet

Live C++ Course on YouTube

Following the trend 🫣

Open AI just launched Dots

3 steps to crack product based companies

The resume that got me into Google

POV: You found real meaning in tech💀 #tech #backend #developer #api #code

Justise for Backend Developers 🥲🥀 #Developers #backend #sedlife

Developers are always right! 💀 #developer #developerlife #sleepcycle

Developers > Testers 💀

What is your favorite soft drink?

Salaries of Gen AI Developers

Complete Gen AI Roadmap #ai #generativeai

It will always feel special to hold the Tricolour beside my father on this day. 🇮🇳❤️ I’ve grown up hearing him say “Jai Hind” on his calls, and even today, years after his retirement, he still says it with the same pride. He served in the Indian Army for 24 years and 23 days. Naib Subedar Jabir Bakhsh — my father, my inspiration, my number one idol and my hero. ❤️ Happy Independence Day! 🇮🇳 Jai Hind! 🇮🇳

Bank of Baroda’s data got leaked

Mummy ke hath ki briyani 😋

My self sustained house in village

Learning from those who inspire millions Alakh Pandey sir has changed how education reaches students across India. Grateful for this moment.

#physicswallah❤️ #rajwantsirphysicswallah #pankajsirphysicswallah #sachinsirphysics #fraz

Next step after JEE

Desi coder mindset 😎

Kya aap bhi confused ho ?

🛠️ How I Built SIR Data Digitizer — Step by Step The government releases Special Summary Revision (SIR) data as PDF files. These are raw, unstructured, and nearly impossible to work with at scale. So I built a tool that digitizes them into a clean, structured CSV — here’s the full process: Step 1 — The Raw Idea Spotted a real problem: SIR PDF data is unreadable at scale. People need structured voter lists for specific polling booths — so I decided to build a digitizer. Step 2 — Planning with Claude Sonnet 4.6 (Thinking Model) Used Claude’s thinking model to plan the complete architecture — the data flow, tech stack, output format, and edge cases — before writing a single line of code. Step 3 — Building with Antigravity (Vibe Coding) Took the plan directly into Antigravity and built the entire tool through structured AI prompting. Prompt by prompt, the full application was shaped exactly as envisioned — no manual boilerplate. Step 4 — OCR Engine: Gemini 2.0 Flash The AI reading the PDFs is Google’s Gemini 2.0 Flash. The SIR data is in Hindi, and Gemini outperforms most other models — including Claude — when it comes to accurately identifying and structuring Hindi text. Step 5 — JSON Preview Once Gemini processes the PDF, the extracted data is displayed as a live JSON preview on the platform so users can verify it before exporting. Step 6 — CSV Export One click converts the JSON into a downloadable CSV — every name, every entry from the government PDF, fully structured and ready for booth-level analysis. Step 7 — Deployed on Google Firebase The full-stack application is live on Google Firebase — handling frontend, backend, and hosting all in one place. Step 8 — Testing Validated with real government PDFs. The experience is intentionally simple: drop a PDF → get clean, structured data. No technical knowledge required.

Why companies regret firing due to AI ?

Why food courts are on top floor in mall?

Love life ❤️‍🩹