Aakash Dabas

• Part of Research and Development Team
• Providing Machine Learning and Data Analytics Expertise within team.
• Creation, modeling and designing end to end systems with use of data science and machine learning algorithms
• Led initiatives in creating state of art proof of concepts which then developed into real life products
• Managed bootstrapping of various data science products
• Worked with project managers, developers, quality assurance and customers to resolve technical issues.

PROJECTS

1. Think-Rf Signal Analysis

• Designed Data management and data caching module for capturing of spectral Data
• Created aws Greengrass module for node to cloud data transfer with autonomous data cleaning system
• Created data models on aws platform to analyze captured data

2. Anomaly Based Intrusion Detection System

• Designed project approach and led project activities
• Built and Deployed machine learning models which could identify and isolate cyber attack in real time
• Built analytics pipeline which was able to collect network statistics for analysis
• Built analytics dashboard which was able to display real time findings and networks to end user.

3. DRDO Gps Navigation System

• Designed data collection and management system of gps data for DRDO
• Created data collection and analysis pipeline for gps data
• created on sensor data management module which can be implemented on any soc.

4. Asset Management System Using Image Tokenization

• Bootstrapped entire project and led design and architecture of project
• Designed novel image tokenization approach using visual transformers
• Wrote machine learning based metadata generation system which can be used to identify files in real time
• Designed secure enclave approach for token storage
• Trained and deployed steganography based ml model which was able to hide one image onto another without any visual discrepancy

5. Quantum machine based spam and deep fake detection

• created a quantum based text classifier which is able to detection text spam in real time
• Created quantum based identity classifier based on user personal data

• Created data visualization graphics, translating complex data sets into comprehensive visual representations.
• Identified, analyzed and interpreted trends in complex data sets using supervised and unsupervised learning techniques.
• Leveraged Statistical techniques to create engineering solutions