Naresh User

I’m an AI-focused Data Engineer with 3+ years of experience building scalable ETL/ELT pipelines, real-time streaming systems, and cloud-native data platforms. I focus on performance optimization across Spark/Snowflake, with proven impact in reducing latency, improving data quality, and enabling self-service analytics.

In AI/LLM data systems, I built and maintained PySpark pipelines on AWS (S3, Glue, EMR), added Airflow orchestration with retry logic and SLA alerting, and reduced pipeline failures by ~15%. I deployed Kafka + Spark Structured Streaming for real-time sensor ingestion, optimized Snowflake models (query profiling, clustering keys, materialized views) to cut dashboard runtimes by 25–30%, and enforced data quality with Great Expectations.

I also delivered hands-on LLM-enabled data applications, including an LLM-powered NL-to-SQL interface using LangChain, OpenAI APIs, and FAISS to reduce manual SQL dependency by ~30%. Alongside production work, I’ve built RAG-based assistants (Text-to-SQL), real-time streaming anomaly detection pipelines, and semantic document search/Q&A systems with FastAPI and Docker.