Bharath Reddy Nallu

About

Data/ML engineer in financial services (Northern Trust) who built a production RAG-based LLM system to connect structured transaction/portfolio data with unstructured market and internal documents for risk teams. Strong in end-to-end pipelines (AWS Glue/Airflow/PySpark), entity resolution, and taking models from prototype to reliable daily production with performance tuning (LoRA + TensorRT) and monitoring.

Experience

Education

Key Strengths

• Built production RAG pipeline linking structured financial data with unstructured reports/notes for risk workflows
• Designed entity resolution combining deterministic ID matching, fuzzy name matching, and embedding-based similarity
• Validation-driven approach using labeled data and threshold tuning to prioritize precision for risk metrics
• Performance optimization of LLM system via LoRA fine-tuning and TensorRT to meet daily-use latency needs
• Production-grade data engineering practices: modular/config-driven code, idempotency, backfills, schema checks, logging/alerting, CI/CD
• Experiment-to-production decision-making using MLflow tracking and backtesting stabilization criteria
• Ongoing monitoring for drift via match-rate and anomaly-pattern tracking

Reference Highlights

• Strong ability to design ML solutions at scale
• Builds reliable and efficient systems
• Improved model prediction accuracy
• Reduced financial risk
• Strong Python expertise
• Effective at automating ETL and data integration
• Excellent communication skills
• Translates complex model results into actionable insights for executives
• Works closely with risk compliance and DevOps teams
• Takes full accountability for problems
• Independently manages end-to-end workflows
• Evaluates multiple optimization approaches
• Would rehire; viewed as a great asset