Mustansir G.
Full-Stack Software Engineer
Cybersecurity-aware Fullstack Engineer for Early-Stage Companies
ABOUT
As a Startup Fullstack Engineer, I facilitate seamless collaboration and strategic alignment between technical and non-technical teams (Founders/Product) to ensure product clarity and drive informed decision-making.
Leveraging my experience, I make calculated and critical decisions concerning technology and core architectural components, while also building and executing these decisions to deliver essential technical processes and infrastructure that underpin product stability and growth.
My expertise enables me to strategically identify and implement the critical features that empower startups to evolve from initial prototypes (0.0.1) to scalable, production-ready products (1.0.0).
EXPERIENCE
Oct 2024 - Present As the Tech Lead at Engaze, I play a key role in product ideation and development. Not only do I write code, but I also document features, flows and how users interact with the defined features, alongside the product team.
- Software Architecture
- Team Management
- Documentation
Apr 2024 - Present At CyferWall, I'm trying to innovate and build cyber security products that can help other companies scale and integrate cyber security via CyferWall's SAAS offering.
- Product Design
- Market Research
- Product Lifecycle Planning
Oct 2023 - Mar 2024 Mentored a granted project, funded via GoI (Government of India), worth 15 lakhs to protect against deepfake attacks. Led a team of 4 junior research fellows to create 9 research papers and 1 patent (granted). Launched a multi-modal deepfake prevention solution called TrustTrace.
- Project Management
- Research and Analysis
- Machine Learning
Jun 2023 - Nov 2024 Full Stack Engineer - Dynavate TechnologiesAs a software development agency, Dynavate helped brands like BoultAudio achieve its dream of creating a solo wearable android app to compete with Chinese apps providing sketchy data privacy. Capable of connecting to multiple watch models, based on different chipsets and operating systems, it's a one of a kind app in the wearable ecosystem. App was built to scale to 1,000,000+ users, and provide a better user experience over that offered by the Chinese apps.
- Android
- Wearables
- Scalable Systems
Apr 2023 - Jun 2023 Launched BeTimeFul 3.0 mobile app for iOS and android within 3 weeks of joining. Handled development of the chrome extension, MacOS app, android and iOS apps. Built and integrated custom payment solutions to increase user retention & user acquisition.
- Mobile Apps
- Chrome Extension
- Payments Integration
Aug 2022 - Aug 2023 Engineered a remote-monitoring and management solution for Linux devices from the ground up. Built on completely free and open source software. Integrated Wazuh for security monitoring. Some features include Remote SSH, Parallel Control, 24x7 Monitoring.
- Remote Monitoring
- Open Source
- Security
Jan 2023 - Apr 2023 Designed 10+ CTF's for the Global CyberPeace Challenge event. CTF domains included web, malware, cryptography, buffer overflow, forensics.
- CTF Design
- Cybersecurity
Jul 2022 - Aug 2022 Worked as a backend development engineer focusing on systems such as task processing, Celery. Managed deployment infrastructure, dockerisation. Built and enhanced the authentication platform.
- Backend
- Celery
- Docker
Jan 2022 - Feb 2022 Wrote internal training documents used to brief analysts on cyber security vulnerabilities and reports.
- Technical Writing
- Cybersecurity
Sep 2021 - Mar 2022 Learned cyber security and basic penetration testing.
- Penetration Testing
- Cybersecurity
EDUCATION
M.Tech in Cyber Security –National Forensic Sciences University ( Gandhinagar )
Focused on software engineering, algorithms, and web development. Participated in hackathons and student tech organizations.
Bachelor of Engineering in Cyber Security –Mumbai University ( Shah and Anchor Kutchhi Engineering College )
Focused on software engineering, algorithms, and web development. Participated in hackathons and student tech organizations.
Projects
LinuxAdmin
Remote monitoring and management platform for Linux devices.
- React
- Node.js
- Electron
- Linux
- WebSockets
- Tailwind
RSH
Quick solution for starting reverse shell listeners using multiple tools like nc, sc, etc.
- Python
- Shell
- Open Source
- CLI
Boult One
Cross-platform mobile application to connect with moyoung and coding sdk based wearable devices.
- Flutter
- Dart
- Bluetooth
- Wearables
- Android
- iOS
Netaji
A poster making application built using a cross-platform language.
- Flutter
- Dart
- Canvas
- Cross-Platform
TrustTrace
Advanced deepfake detection solution for video, audio and images. 10+ SCOPUS indexed papers published, 1 patent published.
- Python
- Deep Learning
- TensorFlow
- OpenCV
- Research
BoultAudio Shopify
Shopify store for BoultAudio, one of India's fastest growing wearables brand.
- Shopify
- Liquid
- JavaScript
- E-commerce
BeTimeFul Mobile
An application meant to improve productivity by automatic blocking of applications.
- Flutter
- Dart
- Android
- iOS
- Productivity
BeTimeFul Web
A Chrome extension meant to improve productivity by restricting access to sites like Instagram, YouTube and more.
- JavaScript
- Chrome Extension
- Web
- Productivity
WORKS
- Patent
A deep-fake detection system and method for detecting potential misinformation spread through manipulated content
Dr. Nilakshi Jain, Maj. Vineet Kumar, Dr. Bhavesh Patel, Dr. Shwetambari Borade, Mr. Mustansir Godhrawala, Mr. Shubham Kolaskar, Mr. Yash Nagare, Mr. Pratham Shah, Mr. Jayan Shah — Patent No. 18/2024, 2024
Filed: Apr 10, 2024
A comprehensive system for detecting potential misinformation spread through manipulated multimedia content (video, audio, image). The system integrates advanced machine learning models for deepfake detection, assigns unique case IDs, provides secure…
- Deepfake Detection
- Multimedia Forensics
- Machine Learning
- Audio Analysis
- Video Analysis
- Image Analysis
- Cybersecurity
- EfficientNet
- ResNext50
- MFCC
- SVM
- Case Management
- Research Paper
ResNet50 DeepFake Detector: Unmasking Reality
Shwetambari Borade, Nilakshi Jain, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — Indian Journal of Science and Technology, 2024
This research presents a robust approach for detecting video deepfakes using ResNet50, enhanced with novel image scraping techniques to minimize errors and improve prediction accuracy. The model, trained on CelebDF and FaceForensics++ datasets, achie…
- Deepfakes
- Deep Learning
- GAN
- ResNet50
- FaceForensics++
- CelebDF
- Research Paper
Enhancing Audio Deepfake Detection using Support Vector Machines and Mel-Frequency Cepstral Coefficients
Nilakshi Jain, Shwetambari Borade, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — Journal of Harbin Engineering University, 2024
This paper presents a machine learning system designed to differentiate real from synthetic speech using a Support Vector Machine (SVM) classifier. Trained on the 'for-original' Fake-or-Real (FoR) dataset, which consists of over 195,000 genuine and c…
- Deepfake Detection
- Mel-Frequency Cepstral Coefficients (MFCCs)
- Support Vector Machine (SVM)
- Audio Analysis
- Feature Extraction
- Media Manipulation
- Scalability Challenges
- Responsible Technology Deployment
- Ethical Considerations
- Real-world Applicability
- Research Paper
Harmonizing Algorithms: An Approach to Enhancing Audio Deepfake Detection
Shwetambari Borade, Nilakshi Jain, Bhavesh Patel, Vineet Kumar, Yash Nagare, Shubham Kolaskar, Jayan Shah, Pratham Shah, Mustansir Godhrawala — International Journal of Intelligent Systems and Applications in Engineering (IJISAE), 2024
This research enhances audio deepfake detection by developing a real-time, highly accurate methodology that addresses technological and ethical gaps. The study integrates an MFCC-based SVM classifier (97.28% accuracy) and a Neural Network with attent…
- Audio Deepfake Detection
- Comparative Model Verification
- Ethical Audio Forensics
- Real-time Speech Authenticity
- SVM-Neural Network Fusion
- MFCC
- Attention Mechanisms
- Digital Media Integrity
- Research Paper
Advancements in Video Deepfake Detection: Integration of ResNet50, EfficientNetB7, and EfficientNetAutoAtt B4 Models
Shwetambari Borade, Nilakshi Jain, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — International Journal of Intelligent Systems and Applications in Engineering (IJISAE), 2024
This study presents a novel approach to deepfake video detection by integrating three state-of-the-art models: ResNet50, EfficientNetB7, and EfficientNetAutoAttB4. The ensemble model, trained on diverse datasets, demonstrates superior performance in…
- Convolutional Neural Networks
- Deepfakes
- Deep Learning
- EfficientNet
- GAN
- ResNet50
- Video Forensics
- Model Ensemble
- Knowledge Distillation
- Digital Media Integrity
- Research Paper
Improving Deepfake Audio Detection: A Support Vector Machine Approach with Mel-Frequency Cepstral Coefficients
Shwetambari Borade, Nilakshi Jain, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — International Journal of Intelligent Systems and Applications in Engineering (IJISAE), 2024
This paper presents a machine learning system designed to differentiate real from synthetic speech using a Support Vector Machine (SVM) classifier. Trained on the 'for-original' Fake-or-Real (FoR) dataset, which consists of over 195,000 genuine and c…
- Audio Analysis
- Deepfake Detection
- Feature Extraction
- Media Manipulation
- Mel-Frequency Cepstral Coefficients (MFCCs)
- Support Vector Machine (SVM)
- Research Paper
Deepfake Technology and Image Forensics: Advancements, Challenges, and Ethical Implications in Synthetic Media Detection
Nilakshi Jain, Shwetambari Borade, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — International Journal of Intelligent Systems and Applications in Engineering (IJISAE), 2024
This comparative analysis delves into the dynamic landscape of deepfake technology and its intricate relationship with image forensics. Focused on advanced machine learning methodologies such as autoencoders, GANs, and CNNs, the exploration reveals b…
- Deepfake Technology
- Detection Methods
- Ethical Concerns
- Image Forensics
- Machine Learning Methodologies
- Research Paper
Detecting Deepfakes: Exploring Machine Learning Models for Audio, Video, and Image Analysis
Nilakshi Jain, Shwetambari Borade, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — International Journal of Intelligent Systems and Applications in Engineering (IJISAE), 2024
This study investigates the field of deepfake detection with an emphasis on the use of machine learning techniques in image, video, and audio analysis. It compares the effectiveness of models such as Random Forests, Gradient Boosting Machines, Suppor…
- Deepfake Detection
- Machine Learning Techniques
- Audio Analysis
- Video Analysis
- Image Analysis
- Ensemble Learning
- Performance Analysis
- ResNet
- Random Forest
- SVM
- CNN
- Neural Networks
- Research Paper
Deepfake Detection Using EfficientNetB7: Efficacy, Efficiency, and Adaptability
Nilakshi Jain, Shwetambari Borade, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — International Journal of Intelligent Systems and Applications in Engineering (IJISAE), 2024
This research explores the efficacy of EfficientNetB7, a state-of-the-art Convolutional Neural Network (CNN) architecture, for detecting deepfake videos. The study investigates EfficientNetB7's ability to efficiently discern subtle visual cues indica…
- Deepfakes
- Convolutional Neural Networks
- EfficientNetB7
- Deepfake Datasets
- Machine Learning
- Video Analysis
- Compound Scaling
- Neural Architecture Search
- Research Paper
Securing Visual Integrity: An EfficientNetB4-Based Solution with Attention Layers and Siamese Training for Face Manipulation Detection in Videos
Nilakshi Jain, Shwetambari Borade, Bhavesh Patel, Vineet Kumar, Mustansir Godhrawala, Shubham Kolaskar, Yash Nagare, Pratham Shah, Jayan Shah — International Journal of Intelligent Systems and Applications in Engineering (IJISAE), 2024
This research addresses the challenge of detecting face manipulation in videos using an EfficientNetB4-based model enhanced with attention layers and Siamese training. The proposed approach leverages convolutional neural networks (CNNs) to identify s…
- Convolutional Neural Networks
- Deepfake
- Digital Media Forensics
- EfficientNetB4
- Face Manipulation Detection
- Attention Mechanisms
- Siamese Networks
- FaceForensics
- DFDC
- Video Analysis
MISCELLANEOUS
GDSC Volunteer
DevFest, MumbaiHacks, 2024
Volunteered as a mentor for the Google Developer Student Clubs (GDSC) at DevFest and MumbaiHacks.
Winner, 1st Place (₹30,000 prize)
KJSCE Hack 7.0, 2023
Built a social-media application for recruiters and IT professionals, and took home 1st place with a ₹30,000 award.
Technical Coordinator
CCTNS 2023 Delhi (National Crime Records Bureau), 2023
Served as Technical Coordinator for the Crime and Criminal Tracking Network and Systems (CCTNS) hackathon held at NCRB headquarters in Delhi.
Top 10 in 10+ Hackathons
CodeShastra, Lines of Code, SIH, etc., 2023
Placed in the top 10 across 10+ hackathons, including CodeShastra, Lines of Code, and the Smart India Hackathon.