Hacker
Sahil Rane
@sahil_rane
CSE (Data Science) student at VCET '28. Building full-stack apps, applied AI tools, and engineering software. Passionate about UI/UX, scalable backend systems, and open-source code under @CyberCodezilla.
Elsewhere
Stack
- TypeScript
- Python
- JavaScript
- Java
- Go
- C
- C++
- PHP
- SQL
- PostgreSQL
- Tomcat
- GitHub Copilot
- Vercel
- Render
- React
- Expo
- Kotlin
HeatFlood Guardian: Climate Aware Navigation for Heat and Flood Resilience
Environmental HacksOct 2026
ENV-entors@yashhh_23,@anurag1
Rescue-Link: Offline First Disaster Emergency Response & Tactical AI Command Platform
First CommitSep 2026
AWSome@anurag1,@yashhh_23
PRIVACY SHIELD
The Zerops ChallengeAug 2026
PrivacyShield is an invisible, real-time security guard for artificial intelligence. It automatically detects and masks sensitive personal data, API credentials, passwords, and private certificates before they ever leave your browser or backend—allowing developers, enterprises, and everyday users to harness the power of ChatGPT, Claude, and Google Gemini with absolute peace of mind and zero risk of data leaks. ────── ### The Problem It Solves Every day, millions of software engineers, analysts, and teams copy-paste server crash logs, code snippets, support tickets, and financial documents directly into AI chat windows. Unknowingly, they expose critical API keys, database passwords, credit card numbers, and private customer records to third-party AI servers. Once transmitted, this confidential data can be logged, audited, or retained by external systems beyond your control. ────── ### The Solution: Universal Zero-Trust Protection PrivacyShield sits quietly as an intelligent proxy gateway and universal Chrome extension between your input and AI platform servers. When you type, paste text, or upload a screenshot into any AI platform: 1. Instant Leak Interception: PrivacyShield scans your input in less than 2 milliseconds across 22+ high-sensitivity pattern rules (including OpenAI keys, AWS credentials, database strings, RSA keys, Aadhaar/PAN cards, and Hinglish secret jargon). 2. Interactive Threat Alert: A sleek warning modal overlay alerts you prior to transmission, highlighting exactly which secrets were detected. 3. Smart Token Redaction: Exposed values are replaced with safe, immutable placeholders (such as [DATABASE_URI_REDACTED] or [EMAIL_REDACTED]). 4. Cultural Reasoning Preservation: Rather than performing robotic translation, PrivacyShield injects native context instructions so AI models respond naturally in English, Hindi (Hinglish), or Marathi (Minglish) without altering technical logic. ────── ### Core Key Features • Universal AI Platform Integration: Natively supports ChatGPT, Anthropic Claude, Google Gemini, Perplexity AI, and DeepSeek. • OCR Screenshot Redaction: Automatically extracts and redacts text inside uploaded screenshots before sending them to the cloud. • Zero-Data Retention Policy: PrivacyShield never stores unmasked raw payloads. All inspection occurs in-memory with strict zero-knowledge isolation. • Synchronized Audit Ledger: Every sanitized payload generates a unique verification transaction link (?txId=tx_.. .) for real-time compliance reporting. • Cloud-Native Performance: Deployed on Zerops for high- availability sub-millisecond execution. ────── ### In Simple Words │ "PrivacyShield is like digital bulletproof glass between │ your private data and public AI models. You get all the │ intelligent answers from AI, without giving away your │ secrets."