cv

Software Engineer

Google

Contributed to Security Health Analytics and Cloud Security Insights, building scalable backend infrastructure and customer-facing security capabilities.

• Delivered Simulation APIs for Security Health Analytics' Infrastructure as Code (IaC) Validation feature to detect policy violations in customer test assets. Received a spot award and Customer Empathy award for fast delivery.
• Contributed to backend infrastructure for Cloud Security Insights enabled by Security Graph, helping design, build, productionize, and launch the product from scratch. Awarded several peer awards.
• Designed, implemented, and scaled scheduling functionality for regular cloud-asset vulnerability scans at configurable intervals, including stopping execution for offboarded customers. Onboarded around 1300 customers without a single scheduling error and received a spot award.
• Implemented, productionized, and scaled a custom query editor enabling free-form semantically correct queries for security-related cloud infrastructure context; also improved internal on-call debugging workflows.

Research Assistant

Systems for Artificial Intelligence Lab, Georgia Tech

Worked on model checkpoint compression and federated learning research.

• Worked on Inshrinkerator, a novel model checkpoint compression algorithm that traverses the compression and accuracy tradeoff space; achieved up to 39x compression and withstood up to 10 restores with negligible accuracy impact in fault-tolerant training.
• Created a novel Federated Learning methodology performing network-adaptive and distributed pruning to better utilize variable per-client network conditions while maintaining comparable accuracy to the baseline model.
• Advisor: Prof. Alexey Tumanov. Inshrinkerator paper presented at ACM SOCC, 2024.

Research Assistant

Entertainment Intelligence and Human-Centered AI Labs, Georgia Tech

Worked on computational creativity and multimodal AI research.

• Contributed to Creative-Wand, an AI co-creativity transformer-based agent that generates stories guided by user goals and input. Won the best paper at ICCC, 2023.
• Worked on hate speech understanding using multimodal attention networks, including PCA on 2048-dimensional ResNet features, computer vision preprocessing and analysis, ensembling, and GradCAM attention visualization.
• Advisor: Prof. Mark Riedl.

Software Engineering Intern

Google

Worked on mitigating storage and compute costs in the Play Analytics backend.

• Designed and implemented a multithreaded concurrent data classification and filtering RPC service that was debuggable and could be enabled or disabled, saving 10% of total storage costs.
• Awarded a return offer to join full-time.

Software Development Engineer II

Adobe Systems, India

Contributed to production and maintenance of backend services for Adobe's Digital Marketing product Campaign.

• Worked on backend services on PostgreSQL, written in C++ for Adobe's Digital Marketing product Campaign.
• Delivered independent features for a product used by over 3k customers, collaborating across product management, UI design, and engineering teams.
• Contributed to the Reporting service for Product Campaign, used by 35K customer domains and extensively coded in Java.
• Identified and reduced security vulnerabilities in a Kafka-based framework using Python, helping uphold GDPR compliance and ensuring zero compliance breaches for European customers.

Software Development Engineer

Adobe Systems, India

Worked on backend and UI technology modernization for Adobe Digital Marketing products.

• Contributed to migration of Adobe Target's colocation project from Apache Sling-based framework to Spring Boot, mitigating bottlenecks and decreasing backend API response time.
• Delivered a proof of concept for compatibility between Backbone and React UI components in Target's UI for an upcoming technology stack upgrade.

Software Development Intern

Samsung Research Institute

Created an image-processing-based Tizen application GUI and optimized backend functionality, reducing backend API response time.

• Awarded a return offer to join full-time post graduation.

Master of Science

Georgia Institute of Technology

Computer Science

• Systems for ML
• NLP
• ML
• Big Data Systems
• AI for Storytelling
• GameAI

Bachelor of Engineering

Netaji Subhas Institute of Technology, University of Delhi

Information Technology

Customer Empathy Bronze Winner

Google

Awarded during tenure at Google.

Spot Bonuses

Google

Three spot bonuses during tenure at Google.

Peer Bonuses

Google

Seven peer bonuses and one kudos during tenure at Google.

ACM SOCC Session Chair

ACM Symposium on Cloud Computing

Invited to chair a session and moderated a session with six research talks.

Narotam Sekhsaria Foundation Scholarship

Narotam Sekhsaria Foundation

Awarded to 15 applicants from 6000+ nationwide in India to pursue higher education.

Research and Patent Award

Delhi University

Granted for exceptional research contributions to the top 5 undergraduate researchers from a batch of 1700 students.

Bootcamp Hackathon Winner

Adobe

Winner among a batch of 80 New Graduate Employees at Adobe.

IBM National Challenge Winner

IBM Research Showcase, IIIT Delhi

Created an IBM Watson-powered Banking Assistant Chatbot that beat 10 other teams in a nationwide hackathon.

Merit Scholarship

Delhi University

Awarded to one of the top 5 meritorious students in a batch of 300 for undergraduate education.

Inshrinkerator: Compressing Deep Learning Training Checkpoints via Dynamic Quantization

ACM Symposium on Cloud Computing

A. Agrawal, S. Reddy, S. Bhattamishra, V. Nookla, V. Vashishth, K. Rong, A. Tumanov.

Beyond Prompts: Exploring the Design Space of Mixed-Initiative Co-Creativity Systems

International Conference on Computational Creativity

Z. Lin, U. Ehsan, R. Agarwal, S. Dani, V. Vashishth, M. Riedl.

RLProph: a dynamic programming based reinforcement learning approach for optimal routing in opportunistic IoT networks

Wireless Networks, Springer

D.K. Sharma, J.J.P.C. Rodrigues, V. Vashishth, A. Khanna, A. Chhabra.

GMMR: A Gaussian Mixture Model Based Unsupervised Machine Learning Approach for Optimal Routing in Opportunistic IoT Networks

Computer Communications, Elsevier

V. Vashishth, A. Chhabra, D.K. Sharma.

A Machine Learning Approach Using Classifier Cascades for Optimal Routing in Opportunistic Internet of Things Networks

IEEE International Conference on Sensing, Communication, and Networking (SECON)

V. Vashishth, A. Chhabra, D.K. Sharma. Presented at Northeastern University, Boston, USA.