Saurav Joshi Email & Phone Number

Chicago, Il, Us

Sunnyvale, California, United States

1. Developed a Packet Injection Toolkit: Designed a highly customizable packet injection tool for the EVO Operating System, enhancing testing processes and reducing packet flow analysis time by 40%. This tool facilitated ASIC-level validation of PTX router configurations, streamlining verification and debugging workflows.2. Implemented Route Information Dump Script: Automated the extraction of next-hop IDs and tokens for specific trap codes, significantly cutting down the daily debugging time by 35 minutes per team member. This automation improved overall team efficiency and accuracy in handling complex routing tasks.3. Optimized GDB Macros for Advanced Forwarding Toolkit (AFT): Crafted advanced GDB macros to debug AFT structures, resulting in an 80% reduction in core dump debugging time. This optimization greatly accelerated the troubleshooting process and enhanced productivity during critical debugging phases.

Chicago, Illinois, United States

1. CloudSweeper Development: Engineered a full-stack Google Drive optimization tool utilizing WebtreeMaps, DataGrids, and document embedding analysis, leading to the identification and deletion of redundant files. This initiative freed up to 22% of drive space and improved data management efficiency by 30%.2. Optimized Data Processing Structures: Re-architected a tree-based storage structure to enhance data aggregation speeds by 50%, significantly improving overall system efficiency and streamlining data handling processes.3. Containerized Deployment and Microservices Orchestration: Developed scalable microservices for real-time data scraping and prebid analysis, reducing latency by 18% and boosting performance by 30%. This setup enabled efficient parallel data collection across multiple websites, enhancing the speed and accuracy of data processing.4. Ad-Retargeting Security Research: Conducted in-depth security research on ad-retargeting, exploring vulnerabilities and identity entanglement. Demonstrated how internet identities could be exploited using first-party cookies, highlighting potential risks and mitigation strategies in digital advertising.

1. AIShiksha Platform Transformation: Collaborated with Amazon Web Services (AWS) to scale the nationwide AIShiksha e-learning platform by transitioning from a monolithic architecture to containerized microservices. This strategic shift led to a 15% increase in user traffic and significantly enhanced the platform's scalability and performance.2. Real-time Communication Systems: Led the backend development and AWS migration efforts, employing Node.js, Python, MongoDB, and Socket APIs to ensure high availability and real-time communication. These enhancements significantly boosted platform adoption and improved overall system performance.3. Smart SageMaker Integration: Designed and implemented an automated user notification and idleness handler system for Amazon SageMaker notebooks, reducing operational costs by 65%. This solution optimized resource allocation and usage, improving efficiency and reducing idle resource costs.4.Microservices Architecture: Developed robust backend services utilizing Node.js and Python, leveraging Docker for containerization and Kubernetes for orchestration. This approach improved system reliability and enabled automatic scaling based on user demand.5. Database Optimization and Scaling: Migrated legacy SQL databases to MongoDB with replica sets and sharding, enhancing query performance by 70% and reducing load times under heavy user traffic. This migration ensured high availability and boosted overall system responsiveness.6. Asynchronous Task Processing: Integrated AWS SQS and Lambda to handle asynchronous tasks, reducing API response times by 65% and enhancing the user experience during peak load periods.7. RESTful API Development: Engineered and optimized REST APIs using Node.js and Python, improving data exchange efficiency and ensuring seamless integration with frontend services.

Bengaluru, Karnataka, India

• 𝗛𝗶𝗻𝗱𝗶 𝗔𝗦𝗥: Pioneered a Hindi Automated Speech Recognition Model, establishing a technology benchmark with the Kaldi Speech toolkit and deep learning techniques. This effort entailed extensive data collection, training, and fine-tuning, achieving high accuracy in Hindi speech processing.• The project’s significance was underscored by a publication at an IEEE conference, marking a significant advancement in speech recognition research and setting a new industry benchmark with a 10.6% Word Error Rate (WER).

Gurugram, Haryana, India

Master'S Degree, Computer Science, 4.0/4.0

1. Parallel and Distributed Processing 2. Secure web application development 3. Algorithms 401 4. Computer System Security 5..

Bachelor Of Technology (B.Tech.), Computer Science, 1St Division

Activities and Societies: Head Coordinator for Training & Placement Team, Ablaze Team, KISLAY Team. Drummer for Symphony Re-tuned.

Junior High/Intermediate/Middle School Education And Teaching