Sri Pujitha Bollu
I’m an AI/ML Engineer with 4+ years of experience building production-ready GenAI and analytics solutions, with a strong focus on RAG systems, predictive analytics, and evaluation/observability for quality.
About
I’m an AI/ML Engineer with 4+ years of experience building production-ready GenAI and analytics solutions, with a strong focus on RAG systems, predictive analytics, and evaluation/observability for quality. I’ve worked across end-to-end pipelines—ingestion, feature/model readiness, governed retrieval, and API deployment—so teams can get reliable insights with reduced manual effort and fewer compliance risks.
In my recent role, I architected a document intelligence platform that centralizes enterprise documents and improves policy retrieval precision through better chunking, embeddings, semantic search, reranking, and metadata filtering. I also set up monitoring and evaluation (latency, faithfulness, hallucination signals, token usage, and feedback loops) to continuously improve answer quality for stakeholders across product, compliance, support, and leadership.
- 4 years' experience
- Joined September 2026
Industry experience
- Software & Internet
- Professional Services
- Financial Services
Work experience
Architected a RAG document intelligence platform to reduce manual policy search for multiple internal teams. Led governed ingestion from AWS S3, PDFs, Word, tickets, and knowledge articles, extracting metadata and RBAC permissions across multiple sources. Optimized chunking, embeddings, and semantic search using sentence transformers, Pinecone/FAISS, and PostgreSQL pgvector to improve retrieval precision across document categories. Designed LangChain/LangGraph workflows with metadata filtering, reranking, prompt engineering, and source citations to reduce hallucinations and support compliance needs. Deployed FastAPI REST APIs using Docker, Kubernetes, AWS ECS/EKS, and CI/CD, maintaining high uptime. Implemented evaluation and observability with RAGAS, LangSmith, MLflow, CloudWatch, Prometheus, and Grafana for latency, faithfulness, token usage, errors, and feedback loops, improving answer quality through monitoring and dashboards.
Analyzed customer churn for thousands of accounts using subscription, billing, payment, and support-ticket data with SQL, Python, and Pandas. Built a customer 360 dataset by consolidating customer, subscription, product usage, payments, and ticket data to strengthen churn analysis, renewal tracking, and predictive revenue-risk reporting. Standardized datasets to resolve duplicates, missing values, and date issues to enable model-ready feature engineering. Created dashboards in Power BI/Tableau/Matplotlib for churn, retention, engagement, and payment failures, reducing manual reporting effort. Segmented customers via cohort analysis and trained Scikit-learn churn scoring models to reduce avoidable churn. Presented churn drivers and NLP-based ticket signals to cross-functional stakeholders across customer success, sales, product, and finance to support outreach and renewal planning.
Education
Hire Sri Pujitha Bollu today
Post a job and invite Sri, or contact them directly. No hidden commission.