Headline

Engineering @Zscaler | Ex-Zeta (Directi), ADP | Distributed Systems, Java, Go, Sprint Boot, Kubernetes, AI and LLMs

About

Senior Backend Engineer with 5 years of experience building high throughput, fault tolerant distributed systems at scale. Core expertise in Java, Spring Boot, Kafka, Elasticsearch, AWS, AI, LLMs and Kubernetes. At Zscaler I have delivered several high-impact systems: 🔹 Designed async bulk processing framework handling 10M+ DLP incidents using Kafka directly contributing to onboarding 100+ enterprise customers. 🔹 Independently diagnosed and fixed OOM crashes in a business-critical production service through heap dump analysis eliminating all customer facing timeouts. 🔹 Built AI driven resolution pipeline using LLMs to analyze error logs, generate fix suggestions, and auto-create PRs reducing resolution time from hours to under 25 minutes. 🔹 Reduced pod startup time from 5-10 minutes to 30-40 seconds through lazy loading eliminating service disruption risk during high traffic. Most of my strongest work came from problems I identified independently I take full ownership from problem identification to production delivery.

Location

India

AI stack

cursor

Devin

Work experience

Senior Software Development Enginner at Zscaler, Oct 2024 - Present · 2 yrs, Bengaluru, ➛ Designed and scaled an asynchronous batch-processing engine supporting 1M+ bulk operations, leveraging parallel execution and deferred aggregation to significantly improve throughput. ➛ Eliminated memory bottlenecks by redesigning large-scale file sorting using external merge sort, enabling reliable processing of 1M+ records without OOM failures. ➛ Reduced analytics dashboard load time by 5 seconds by removing dynamic templating and optimizing workflows. ➛ Built an end-to-end AI-driven incident resolution pipeline that analyzes error logs, identifies impacted file paths, retrieves source code from repositories, generates fix suggestions using LLMs, and auto-creates PRs via cron jobs, reducing resolution time to 5 to 25 minutes. ➛ Automated workflow creation by converting customer requirements provided via text or voice into structured, LLM-generated workflow specifications, reducing customer effort by 5 to 10 minutes per request. ➛ Built a high-throughput aggregation service consolidating more than 1 millions of incident records, reducing duplication and improving system reliability. ➛ Created reusable Kafka consumer and producer frameworks, standardizing retry strategies, DLT handling, and extensible stream processing. ➛ Implemented real-time WebSocket pipelines handling 10K+ concurrent requests per minute, ensuring low latency and resilience in Kubernetes-based deployments. ➛ Delivered an enterprise-grade audit logging platform enabling traceability, compliance, and customer transparency.

Software Development Enginner 2 at Zeta, Jul 2024 - Oct 2024 · 4 mos, Hyderabad, ➛ Automated user and storage provisioning workflows, reducing setup time from 5 days to under 10 minutes, demonstrating ownership and bias for action. ➛ Enhanced observability by integrating Prometheus metrics and Grafana dashboards—increasing visibility from 70% to 95% and enabling data-driven operational insights. ➛ Redesigned the organization-level scheduler to optimize resource allocation and execution throughput, boosting scheduling efficiency by 30% while maintaining reliability at 10k requests/sec scale. ➛ Strengthened customer security measures by transitioning from access keys and secrets to IAM role-based authentication for S3 bucket access. This enhancement bolstered security protocols and minimized potential vulnerabilities. ➛ Designed and implemented a service to facilitate secure access, transfer, and management of large files and sensitive data using the SFTP protocol. This innovation provided customers with a reliable and secure solution for handling their data needs.

Software Development Enginner at Zeta, Oct 2022 - Jul 2024 · 1 yr 10 mos, Hyderabad, Telangana, India

Software Engineer at ADP, Aug 2021 - Oct 2022 · 1 yr 3 mos, Hyderabad, Telangana, India, ➛ Redesigned and implemented an algorithm to fix the missing numbering gaps for production and test clients which increasing the accuracy from 70% to 95% using Java and Spring Boot. ➛ Developed a mechanism to make sure that the scheduled tasks are executed atmost once for a specific time using Spring Boot and Java. ➛ Analysed, redesigned, and implemented the Employee Health Tax Xml file for multiple provinces and increases the provinces for 1 to 4 using Spring Boot, Java. ➛ Created an excel tag processor to fetch data from an excel sheet to enhance the e2e test cases to compare the value from REST API and excel sheet using Groovy Framework. ➛ Strengthened the accuracy of the project while generating Employee Health Tax, PayDisbursement XML files for multiple provinces from 76% to 91% using the Junit framework.

Teaching Assistant at Coding Ninjas India, Dec 2019 - Mar 2020 · 4 mos, New Delhi Area, India, ➛ Worked there as a Teaching Assistant for the course on Data Structures and Algorithms. ➛ Helped students by clearing their doubts, tried improving their problem-solving, ability by providing new and novel problems to them. ➛ Able to successfully resolve more than 200 doubts for various topics in Data Structure & Algorithms.

Campus Chapter Vice President at CodeChef, Nov 2019 - Jul 2020 · 9 mos, Faridabad, Haryana, India, ➛ Helped the juniors to understand the data structure and algorithms and solving problems of various kinds in Leetcode, Codechef etc. ➛ Conducted programming Contests at the University for better to improve their problem-solving skills with a short range of time. ➛ Increased the size of the club from 5 to 53 students.

Education

Bachelor of Technology, Computer Science, Manav Rachna Educational Institutions

Public links

linkedin.com: https://www.linkedin.com/in/vaibhav-vashist/

leetcode: https://leetcode.com/vaibhavvashist9999