Software Engineer – Web Dev • Cloud • DevOps • AI Engineering
- I build back-end services, security automation, and cloud infrastructure with a focus on reliability and scalability.
- Experienced with Python, Flask, FastAPI and Node.js/TypeScript for API development and microservices.
- Proficient in AWS/GCP cloud platforms, Terraform IaC, Docker/Kubernetes, and CI/CD with GitHub Actions.
- Building with LLM APIs, RAG pipelines, and AI/ML integrations for production-grade intelligent applications.
- Interested in observability, security automation, and production reliability engineering.
University of Maryland College Park | College Park, MD
Master of Engineering, Software Engineering | GPA: 3.79/4.0 | Aug 2024 - May 2026
- Teaching Assistant for Computer Vision (Fall 2025, Spring 2026)
- Teaching Assistant for Algorithms and Data Structures for Machine Learning (Spring 2025)
Vellore Institute of Technology | Vellore, India
Bachelor of Technology, Electronics and Communication | GPA: 9.07/10.0 | Jul 2018 - May 2022
- Shipped 20+ REST APIs (Python/Flask) with comprehensive Swagger documentation, resolved 40+ defects, and wrote 80+ PyTest cases → higher reliability & maintainability.
- Built a serverless CVE alerting workflow (AWS Lambda/EventBridge/PostgreSQL) with idempotency → 3-hour detection SLA and zero duplicate notifications.
- Deployed a GitHub App-based secret scanner (Lambda/SQS/Fargate/Gitleaks) with PR checks → ~60% reduction in credential exposure; ~50% less manual review.
- Provisioned EKS via Terraform, deployed OpenTelemetry with Helm, added CI/CD with rollback and Prometheus/Alertmanager → Slack → faster, auditable releases.
- Automated GCP deployments with Docker + Terraform, addressing 100+ Artifact Registry issues → ~40% faster release cycles.
Software Developer Intern | Jun 2025 - Aug 2025
Security Automation & Cloud Reliability
-
CVE Alerting (Serverless Pipeline)
- Built an idempotent serverless pipeline on AWS (Lambda, EventBridge, PostgreSQL) to parse vendor CVEs every 3 hours, eliminating duplicate alerts and ensuring timely Office 365 email notifications.
- Impact: 3-hour detection SLA, zero duplicate alerts, faster routing to security owners.
-
Secret Scanning (Merge Protection)
- Deployed GitHub App-based secret scanner leveraging Lambda, SQS, Fargate, and Gitleaks to securely clone repositories, run scans, and block PR merges containing credentials.
- Impact: ~65% reduction in credential-exposure incidents; ~50% less manual review effort.
-
Automated Notifications & Compliance
- Implemented automated notification workflow with Office 365, instantly alerting developers of detected secrets while enforcing branch protection rules.
- Impact: Improved security compliance by 80% and reduced manual review effort.
Tech: Python, AWS (Lambda/SQS/Fargate/EventBridge), PostgreSQL, GitHub App/Checks API, Gitleaks, Office 365, GitHub Actions
Software Developer I | Jul 2022 – Jun 2024
API Platform & Cloud Engineering
- Built: 20+ Python/Flask REST APIs with comprehensive Swagger/Postman documentation for Company/User details and Escalation services.
- Quality: Wrote 80+ PyTest unit test cases and resolved 40+ critical API bugs, ensuring reliability and stability.
Cloud Release Engineering
- Automated: GCP deployments using Docker + Terraform; addressed 100+ Artifact Registry issues.
- Outcome: ~40% faster release cycles; stabilized build/publish paths.
Observability & Data Engineering
- Delivered: 20+ Looker (LookML) dashboards for SIEM/SOAR workflows.
- Built Python-based solutions for BigQuery connections, Cloud Scheduler jobs, and Cloud Monitoring alert policies for incident detection.
Development Practices
- Standardized: API patterns, test templates, and 50+ POC documentation to accelerate onboarding and code reuse.
- Boosted sprint efficiency by 20% through active contribution to Agile methodologies (sprint planning, retrospectives, user stories, code reviews).
Tech: Python/Flask, SQL, Postman/Swagger, Docker, Terraform (GCP), BigQuery, Looker/LookML, Cloud Monitoring, Git, PyTest
Data Analyst (Aero Data Analytics) | Jan 2022 - May 2022
Digital Twin Analytics
- Developed engine performance data analytics system using MATLAB and Turbomatch, achieving 95% accuracy in predicting engine parameters.
- Simulated various flight conditions to generate key engine parameters (fuel flow, pressure ratio, gross thrust), facilitating optimal maintenance decisions.
Signal Processing
- Implemented scattered interpolation algorithms in MATLAB to process remote engine data, reducing error rates to <1% (Case 1) and ~3% (Case 2).
Tech: MATLAB, Turbomatch, data interpolation, experiment design
- Integrated Claude Vision API and NASA GIBS satellite imagery with FastAPI and pandas to classify transaction-level CO2 emissions, generate reduction recommendations, benchmark against Paris Agreement targets, and detect land-cover changes from Landsat imagery.
- Configured a dual-mode research pipeline using PyPDF2, ChromaDB, and Voyage AI embeddings for RAG-based cited Q&A across indexed PDFs, alongside a separate Claude-powered summarization service extracting findings, metadata, and figure analysis.
- Implemented strategic caching and prompt engineering across three production-ready services to reduce API costs by 70% while maintaining analysis accuracy.
Tech: Python, FastAPI, Claude API (Text + Vision), NASA GIBS, Voyage AI, ChromaDB, PyPDF2, RAG, pandas, Prompt Engineering
- Architected serverless event management platform using AWS Lambda, DynamoDB, AppSync GraphQL API, and EventBridge, supporting 1,000+ concurrent users with RBAC, automated waitlist management, and sub-2-second API response times for 200+ university event organizers.
- Designed event-driven microservices architecture with Node.js/TypeScript backend, Next.js 14 frontend, and CQRS pattern using single-table DynamoDB modeling, achieving 99.5% uptime through multi-AZ deployment.
Tech: AWS Lambda, DynamoDB, AppSync, EventBridge, Node.js, TypeScript, Next.js 14, SAML 2.0, Serverless Architecture, GraphQL
- Developed AI-powered fitness application generating personalized 7-day workout and nutrition plans using Claude API integrated with Flask backend; won Student Choice Award at Anthropic Hackathon.
- Engineered full-stack solution with responsive frontend handling multi-parameter inputs, optimized system prompts for consistent JSON outputs, and implemented age-adaptive exercise programming.
Tech: Python, Flask, Claude API, HTML, CSS, JavaScript
- Built a job watcher with FastAPI, SQLite (SQLAlchemy), Requests + BeautifulSoup, and SMTP to track company career pages, filter postings by keywords, and deliver HTML email summaries; won 2nd Prize in GDG's Hackathon.
- Implemented a minimal UI using HTML, CSS, and JavaScript; prototyped cloud deployment using GCP Cloud Functions + Cloud Scheduler with Secret Manager for SMTP automation.
Tech: Python, FastAPI, SQLite, SQLAlchemy, BeautifulSoup, GCP Cloud Functions, Cloud Scheduler
- Built a packet analyzer decoding Ethernet/IPv4/TCP/UDP and surfacing "top talkers and flows" to pinpoint bandwidth hogs and flaky services.
- Productionized with CMake; added DNS-based domain labeling and TCP handshake metrics (SYN→SYN/ACK RTT, fail rate) to reveal latency/reachability issues.
Tech: C++, libpcap, CMake, Wireshark
- Delivered reproducible observability stack by provisioning EKS with Terraform (VPC, private nodes) and deploying OpenTelemetry via Helm; exposed services with LoadBalancer for Jaeger tracing and Grafana dashboards.
- Implemented GitHub Actions CI/CD to deploy/rollback Helm releases and configured Prometheus + Alertmanager → Slack webhooks with pod-restart alerting rules.
Tech: AWS EKS, Terraform, Helm, OpenTelemetry, Prometheus, Grafana, Jaeger, GitHub Actions
- Engineered a real-time drowsiness detection system using Python, OpenCV, and dlib that tracked eye aspect ratio and lip distance, achieving 90%+ accuracy in detecting yawning and fatigue.
- Integrated pulse monitoring mechanism utilizing Arduino IDE, NodeMCU, and pulse sensor, transmitting real-time data to ThingSpeak cloud, triggering alerts when driver's pulse fell below 60 bpm.
Tech: Python, OpenCV, dlib, Arduino, NodeMCU, ThingSpeak
