Modern data stack
AWS (S3, Athena, Glue, Kinesis, Firehose, SageMaker, Bedrock), Google Cloud Platform (Vertex AI, BigQuery, Google Cloud Build), Parquet, Snowflake, dbt, Databricks, Apache Spark, Alteryx
Data science · Data engineering · Machine learning · Agentic AI
Sneha Nagabhairava
Strategic data engineering professional with over two decades of experience scaling complex platforms and modernizing enterprise data ecosystems, including deploying data solutions at customer utility sites. Proven expertise in Snowflake, dbt, ETL/ELT, SQL optimization, data architecture, and cloud data engineering, alongside hands-on experience developing machine learning, RAG, and agentic AI solutions across AWS and Google Cloud. Focused on scalable pipelines, reliable data transformations, and technical execution.
- Charlotte, NC
- sneha.nagabhairava@gmail.com
- (651) 734-3610
- GitHub
Technical expertise
Skills
AI and machine learning
Large language models (LLMs), machine learning, MLOps, LangChain, agentic AI, deep learning, data science
BI and visualization
Tableau (certified), Power BI, SAP BusinessObjects
Databases
Oracle (certified professional), Postgres, SQL Server, Sybase, NoSQL, ChromaDB, Redis, star/snowflake design
Languages
SQL, PL/SQL, Python, TypeScript, Java (Sun certified)
Tools and platforms
Streamlit, Docker, Apache Airflow, CI/CD pipelines, FastAPI, React, Informatica, Datameer, DBeaver, Toad, GitHub, Jira, Google Workspace, Node.js, ReAct, Microsoft 365
Professional experience
Experience
June 2025 – Present
Applied AI and Machine Learning
Independent
- Built FinanceBuddy, a personal-finance lakehouse with synthetic ledgers, dbt on Athena, ML for subscriptions/habits/fraud, and a Streamlit UI on S3 gold. Optional Kinesis streaming and a Bedrock Haiku coach grounded in marts.
- Built a prototype RAG agent over public Mecklenburg County tax and sales records (Python, LangGraph, Chroma, Streamlit) that routes each question to SQL or document retrieval. Added a Redis caching layer; in local testing, cached repeat queries returned in about 43 ms versus about 8 s uncached (p95).
- Built CivicPocket, a public-assistance finder for Mecklenburg County, with Next.js, FastAPI, Vertex AI, and BigQuery. Common personal details are stripped before a question reaches the model.
- Built a youth mental health risk model on the 2024 National Survey of Children's Health with XGBoost, SHAP, and MLflow. The explainer runs on Streamlit.
Jan 2012 – May 2025
Eagan, MN
Lead Software Engineer (Data Engineering & Analytics)
Thomson Reuters
FindLaw — Performance Analytics Group · 2015 – May 2025
Internet Brands acquired FindLaw in Dec 2024.
- Led cloud data warehouse modernization to Snowflake, designing scalable ELT pipelines to ingest input datasets staged in AWS S3 and orchestrating schedule-driven dbt model runs using Apache Airflow DAGs.
- Championed and implemented dbt from scratch to eliminate redundant, fragmented SQL across 8 analysts, centralizing transformation logic, establishing Git version control, and reducing dashboard development time.
- Decreased data inconsistencies by 35% by creating standardized, certified Tableau data sources used by 20+ analysts company-wide.
- Led the technology transition following Internet Brands’ acquisition of FindLaw from Thomson Reuters by enabling seamless customer data porting and updating 15+ downstream processes to ensure smooth operations after SAP account numbering system restructuring.
- Architected a social media budget reporting solution for Facebook products and an article forecast reporting system for workload balancing.
- Eliminated 15+ hours of weekly manual reporting by designing PPC Budget and Performance Dashboards that resolved complex budgeting issues; received a company-wide Star Performer award.
- Created a Budget Estimator tool that uses logistic regression and random forest classifiers to categorize Google Ads into the correct practice areas, reducing manual effort.
- Prototyped AWS MLOps workflows leveraging AWS SageMaker for model deployment pipelines.
FindLaw — Data Services Group · 2012 – 2015
- Led the BPM fulfillment analytics team and increased offshore team productivity by 28% through structured weekly leadership meetings and clear implementation guidance.
- Reduced query execution time through SQL optimization and data model improvements.
- Performed data modeling, data mapping, and analysis for several key reporting projects such as Oracle BPM reporting, vendor API (crowdsourcing of content), call quality, scheduling of calls, and product performance.
- Integrated FindLaw’s customer data with Thomson Reuters Master Data Management System.
- Implemented several Business Objects universes and developed BO reports.
- Led technical book discussions on advanced analytics and created training materials for the business and developers.
Jan 2007 – Jan 2012
Bloomington, MN
Senior Analyst / Oracle Software Engineer
Ecologic Analytics
- Accelerated data processing by 70% by rewriting core processes with table partitioning and resolving production locking issues.
- Deployed and configured MDMS (Meter Data Management System) at customer utility sites, including Pacific Gas and Electric, Austin Energy, and Indianapolis Power and Light, through on-site client visits and requirements gathering; supported RFP responses. Automated 90% of data migration processes by developing robust ETL frameworks for the Ecologic MDMS platform.
- Enhanced data security protocols for utility customers by designing and implementing comprehensive security architecture including Oracle VPD implementation.
Jan 2004 – Jan 2007
Minneapolis, MN
Database Administrator
Wolters Kluwer Financial Services
- IRAMC database management, build deployments, optimizing SQL queries, and migration.
Aug 2001 – Nov 2003
Mumbai, India
Software Programmer
Infrasoft Technologies
- Reports for Coutts Bank, web frameworks, CMM Level 5 certification, and built database components.
Education
Education
Graduated 2001
Bachelor of Science in Computer Science
Mumbai University, India