7+ years of professional experience. Successfully designed products for Fortune 500 companies. Keen interest in Autonomous aerial robots, Self-driving technologies, Computer Vision and Industrial Big Data Analysis. Adept at developing AI automation scripts for Machine Control and Robotics.
1. Research Data Tools: Conduct thorough research on various data tools available in the market and assess their suitability for different data needs.
2. Build Telemetry Data Pipelines: Design and develop scalable, flexible, reliable, and highly available Vehicle-to-Cloud telemetry data pipelines.
3. Time-Series Data Collection and Storage: Collect time-series data from vehicles and store it in various data destinations (Databases, Data Warehouses, and Data Lakes) to enable real-time observability, long-term storage, and Business Intelligence analytics.
4. Develop Custom Applications and Pipelines: Develop custom C++ applications, Docker scripts, and data pipelines, leveraging tools like Vector, InfluxDB, AWS Timestream, AWS OpenSearch, AWS S3, ClickHouse, Metabase, and AWS Kinesis.
5. Work extensively with gRPC, Protobuf, sockets, and REST APIs to ensure seamless communication and data transfer between distributed systems.
6. Data Migration: Manage data migration between systems (e.g., InfluxDB to AWS Timestream, InfluxDB to AWS S3, AWS S3 to InfluxDB, AWS S3 to ClickHouse) using custom ETL tools for batch processing.
7. Real-Time Vehicle State Management: Develop and maintain the vehicle state agent to send real-time vehicle state snapshots to the Redis server.
8. Grafana Visualization: Build Grafana dashboards for real-time monitoring of vehicle telemetry data which sources from AWS Timestream (InfluxDB)
1. Developed a Django-based SaaS Learning Management System (LMS) product tailored for early school education, handling ~2 TB of content data hosted on an Intranet environment.
2. Designed and built relational models using PostgreSQL to manage user data, course content, scheduling, assessments, and analytics, ensuring efficient data organization and retrieval.
3. Implemented subscription management and membership features, enabling seamless user enrollment and content access based on subscription tiers.
4. Integrated online payment gateways such as Razorpay and Stripe for secure payment processing, supporting various payment methods for subscription and course fees.
5. Built key features such as User Management, Role-based Access Control, Content Management, Course Management, and License Server to ensure scalable and secure content delivery.
6. Integrated advanced functionalities including Data Analytics, Course Scheduling, Admin Portal, Multi-Tenant Support, and Dynamic Assessments to enhance the user experience and learning outcomes.
7. Developed features like Language Change, Recommendation Engine, Content Encryption, Lazy Loading, and HLS Video Chunks to optimize content delivery and provide secure access.
Implemented User Activity Tracking and Learning Objective-based Student Assessments to monitor progress and improve learning paths.
8. Created a Windows Installer for seamless deployment of the LMS product across client machines.
Led the implementation of CI/CD pipelines using GitHub, enabling automatic Docker image creation and deployment to AWS ECS with AWS ECR for container storage.
9. Led program management, provided training and mentorship to team members, and facilitated customer interactions to ensure alignment with business objectives and client needs.
1. Develop FLIR Thermal camera data grabber and integrate with ROS
2. Work on enhancement of Tracking algorithm for dynamic objects around autonomous truck using Extended Kalman Filter
3. Design and develop End-to-End data platform to collect ROS bag data, Vehicle telemetry (Live) and system log data (Live) from autonomous vehicle. Deployed this platform in AWS Cloud. Features of this platform includes Data upload, Data Visualization (ROS data/Vehicle Telemetry), Data Snapshot, Data Parsing (For Annotation), Data Download and Access Management. Reduces data pre-processing and annotation work by 90%.