AI-Driven Software Engineer with Data Pipelines

VARUNKUMAR SONAWANE

*********@*****.*** 812-***-**** linkedin.com/in/varun-sonawane GitHub Portfolio Summary

AI-focused Software Engineer with a growth mindset and hands-on experience building data pipelines and scalable backend systems, with strong data structures, problem solving, analytical reasoning, and cross-functional collaboration, integrating AI agents and LLM-based automation on AWS and GCP using Python and SQL. Education

Indiana University Bloomington, IN, USA. Aug 2024 – May 2026 Master of Science in Computer Science CGPA: 3.83/4 Certifications

AWS Certified Developer – Associate Microsoft Azure AI Fundamentals Experience

Graduate Teaching Assistant, Indiana University, Bloomington, IN Dec 2025 - Present

• Mentored 120+ students on core software engineering topics, providing detailed feedback on code quality, documentation, and object-oriented design.

• Guided teams through the full software development life cycle (SDLC) including requirements, design, implementation, testing, deployment, and debugging of distributed systems, NLP pipelines, and web scraping applications in Python, Scala, and Spark.

• Collaborated with the instructor to develop assignments, project rubrics and coding labs on SQL, version control, Agile workflows, and Tableau dashboards, strengthening end-to-end engineering practices. Data Engineer Intern, The Commons XR, San Diego, CA May 2025 - Aug 2025

• Designed and implemented scalable backend pipelines to migrate and validate 5M+ records per week from PostgreSQL to BigQuery, eliminating manual handoffs and improving system reliability.

• Created and maintained 10+ production features and services in Python, Plotly, and Dash to support real-time and ad-hoc analytics for XR platforms.

• Integrated an LLM-powered automation service on Vertex AI to replace manual post-session analysis, generating structured insights and reducing recurring engineering effort. Software Engineering Intern, CodeClause, India July 2023 - Jan 2024

• Increased backend system reliability by reducing recurring service failures across 5+ microservices by implementing unit tests, improving error handling, and integrating changes through a CI/CD pipeline.

• Delivered a high-impact user-facing feature by building and integrating a React-based frontend with REST APIs, improving feature usability and reducing manual support requests.

• Reduced resolution time by 20% by defining infrastructure using Infrastructure as Code (IaC) and adding structured logging and metrics, enabling faster root-cause analysis. Projects

VoiceLegal-AI Web Extension

• Built a conversational AI system for legal document analysis by integrating a web client, backend services, and Chrome extension, delivering 4.8 faster responses than traditional AI tools for real-time question answering.

• Elevated user experience by integrating Vertex AI (Gemini 2.5) and ElevenLabs voice, delivering explainable natural-language responses through chat and voice. Cloud-Based Web Scraping & Analytics Pipeline Glue, Athena SQL, DynamoDB, Power BI, SNS, Docker

• Automated daily ingestion of 10K+ records using containerized Scrapy jobs on ECS Fargate and Amazon S3.

• Build a cloud analytics pipeline with AWS Glue, Athena, DynamoDB, and Lambda for daily dashboards and alerts. IdeaGenie - AI-Driven Idea Evaluation Platform Luddy Hackathon Winner

• Developed a full-stack AI product by developing a Flask REST API and a React frontend to evaluate and rank top 3 user ideas, enabling automated prioritization and reducing manual review effort.

• Improved decision quality and usability by integrating LLM-based reasoning with an interactive UI, allowing users to adjust parameters and receive explainable, real-time feedback. Technical Skills

Languages: Python, SQL, Bash, HTML, CSS Systems: REST APIs, Microservices, CI/CD, Linux Databases: PostgreSQL, MySQL, MongoDB, GraphQL, Cassandra, Neo4j Cloud: AWS, GCP, Azure, Docker, Jenkins, Kubernetes AI/ML: TensorFlow, Scikit-learn, Pandas, NumPy, LangChain, NLTK Distributed: Spark, Airflow Tools: Git, Excel

Contact this candidate