Sai Nekkanti

About

Built a production "sentinel insight engine" to tame information overload from millions of product reviews and support transcripts, combining Azure OpenAI (GPT-3.5) zero-shot classification with a fine-tuned T5 summarizer to generate weekly actionable product insights. Demonstrated strong MLOps/production engineering by adding drift monitoring with embedding-based detection, integrating REST with legacy SOAP/queue-based CRM via FastAPI middleware, and scaling reliably on Kubernetes with HPA.

Experience

Education

Key Strengths

• Built and deployed an LLM-powered customer feedback insight engine for large-scale unstructured text (reviews + support transcripts)
• Designed a two-stage pipeline (LLM zero-shot classification + fine-tuned T5 summarization) to deliver weekly actionable insights
• Implemented model/data drift detection using embedding distribution monitoring (Word2Vec) with automated retraining alerts
• Integrated modern REST services with legacy CRM systems via FastAPI middleware (SOAP/message queues) without modifying legacy systems
• Scaled production workloads on Kubernetes using HPA to handle unpredictable traffic spikes
• Production-minded AI workflow design with modular components, multi-level testing (unit/integration/canary), and measurable metrics (quality + latency/throughput/error rate)
• Strong stakeholder alignment: translated ambiguous PM goals into measurable success criteria (e.g., 90% accuracy on negative sentiment for top features) using mock outputs