I'm a software engineer with 4+ years of experience building high-throughput backend systems and distributed data pipelines — and more recently expanding into GenAI and LLM integration. I recently completed my M.S. in Artificial Intelligence at San Jose State University (GPA: 3.9), where I built an AI multi-agent research platform combining distributed systems with LLM tooling.

Currently based in San Jose, CA and actively looking for Senior Software Engineer roles focused on backend infrastructure, CI/CD, or AI/LLM integration.

Languages Go Python Java SQL Shell Scripting

Backend & Distributed Systems gRPC / Protobuf RabbitMQ Apache Kafka Redis PostgreSQL Django Spring Boot REST APIs

AI & LLM LLM Tool-Calling (MCP) Prompt Engineering LangChain smolagents RAG PyTorch

Observability Prometheus Jaeger Grafana OpenTelemetry ELK Stack pprof

CI/CD & Infrastructure Jenkins Docker Kubernetes KEDA Bazel AWS (SageMaker, EC2)

Jan 2024 – May 2025

Built an AI Multi-Agent Research Platform — a distributed system for analyzing academic papers using LLM agents.

• Architected a Go-based MCP framework as a standardized tool-calling layer for LLM agents — handled concurrent tool requests using Goroutines, with smolagents on top for agent orchestration
• Applied prompt engineering techniques — chain of thought, few shot, and meta prompting — to improve AI agent accuracy across different paper types
• Built Python/Django backend with Kafka event-driven architecture — decoupled heavy AI processing from main request flow, auto-scaled Docker workers on Kubernetes via KEDA based on Kafka consumer lag
• Set up Jenkins CI/CD pipelines — automated testing, Docker image builds, and rolling deployments across all services

Sep 2021 – Jan 2024

Designed and built the integration layer connecting Dave & Buster's existing systems with Salesforce Cloud for real-time pricing, product configuration, and rewards processing.

• Architected Go microservices handling millions of daily real-time events with sub-100ms response times
• Engineered async task processing using Goroutines and RabbitMQ — decoupled reward syncing from checkout flow, eliminated peak-hour bottlenecks
• Executed pprof performance profiling — identified Goroutine leak causing memory growth under load, resolved with context timeouts — reduced CPU utilization by 25%
• Designed gRPC/Protobuf API contracts for inter-service communication — binary serialization significantly faster than REST/JSON at scale
• Implemented Redis distributed caching with event-based invalidation — reduced latency by 40%, eliminated redundant Salesforce API calls
• Contributed to Jenkins, Docker, and Kubernetes CI/CD workflows supporting automated deployments

Jan 2020 – Sep 2021

Maintained Broadcom's License Management application — a Java Spring microservice used by hundreds of vendors globally for hardware and software license generation.

• Modified existing Java Spring microservices to support new external license generation flow — built mock simulation service to unblock development against unstable external API
• Built Python data pipeline using Pandas and ELK stack — automated vendor compliance reporting, replaced manual database queries with real-time Kibana dashboards
• Diagnosed and resolved API performance issues using Spring Boot Actuator and JVisualVM — identified missing indexes and cold start issues, optimized with composite indexes and warmup scripts — reduced response times by 35%
• Improved observability with Prometheus and Jaeger — set up metrics tracking, distributed tracing, and alerting — significantly reduced MTTR for production incidents
• Hands-on experience with Jenkins CI/CD pipelines across large cross-functional teams

M.S. Artificial Intelligence — San Jose State University (GPA: 3.9/4.0) Conferred Dec 2025 Distributed Systems · Deep Learning · NLP · Recommender Systems

B.E. Computer Science — SGSITS Indore, India (GPA: 3.8/4.0) 2015 – 2019

Master's thesis — built a RAG-based framework for detecting AI-generated audio using LLM inference pipelines. Engineered automated ML pipelines to process unstructured audio data for real-time verification.

Multi-threat detection system using YOLOv8 for real-time data stream analysis. Designed AWS infrastructure via SageMaker supporting 50+ concurrent sessions with 99.9% uptime.