
Senior Engineer – Al & Machine Learning
Nagarro · Jan 2026 – Present
- Owned NLP-based incident routing & duplicate-incident detection across 4M+ ServiceNow & Jira tickets; fine-tuned Transformer models using Python, Pytorch, PySpark & Databricks, achieving 90%+ routing accuracy & reducing manual assignment effort by 60%.
- Engineered anomaly-detection pipelines for 30M+ operational and behavioral records to identify data-quality issues and operational anomalies using SQL, Tensorflow, PySpark, Isolation Forest, and XGBoost; reduced false-positive alerts by 55% and escalations by 18%.
- Architected a change-risk platform using LangGraph, RAG, and tool calling to assess 40K+ production changes; combined change history, dependencies, and knowledge to recommend validation checks, achieving 89% high-risk precision and reducing manual review by 48%.
- Developed an automated CVE remediation platform on AWS & Azure using LangGraph, Ansible & MCP to identify vulnerabilities, assess packages, recommend secure versions & validate fixes; added human approval before patching, reducing remediation time by 65%.
- Optimized multi-cloud LLM inference using FastAPI, MLflow, hybrid retrieval, RAG, and knowledge-base retrieval; applied LangSmith and RAGAS for evaluation and monitoring, reducing serving costs by 55% while maintaining reliability.
- Built a multilingual voice platform for L1 customer support with speech-based resolution, RAG-based knowledge retrieval, intent routing, and human-agent escalation; automated 70%+ of routine interactions and reduced response time by 45%.
- Collaborated with product, design, and engineering teams to translate customer requirements into technical solutions, contribute to architecture decisions, and deliver production-ready ML and RAG features while improving reliability, evaluation, monitoring, and performance.


