Greeshma T

AI Engineer with 4+ years of experience designing, deploying, and scaling production-grade AI/ML solutions and backend systems in fast-paced environments. Specializes in LLM fine-tuning, RAG pipelines, agentic workflows, semantic search, and real-time API development. Proven track record in optimizing latency, reducing compute costs, and delivering intelligent automation in healthcare and enterprise domains.

Proficient in Python, Java, SQL, Docker, AWS, Kubernetes, and MLOps. Collaborative engineer skilled in improving team velocity and enabling data-driven decision-making.

AI Engineer - IBM Watson Health - Cambridge, MA

(2024-09)

• Designed and deployed LLM-based RAG pipelines using Llama 3, LangChain, and Pinecone for clinical note retrieval, improving answer relevance by 35% in healthcare QA tasks.
• Built real-time inference APIs with FastAPI and gRPC, reducing p95 latency from 1.2s to 850ms through async processing and model quantization (GPTQ, AWQ).
• Implemented agentic AI workflows for automated medical coding and PII redaction, achieving 92% accuracy and reducing manual review time by 40%.
• Optimized vector search with hybrid search (BM25 + dense retriever) and reranking using Cohere, boosting top-5 recall by 28%.
• Containerized ML services with Docker and deployed on AWS ECS with Fargate, cutting infrastructure costs by 25% via auto-scaling and spot instances.
• Built ML pipelines using AWS SageMaker Pipelines and MLflow for experiment tracking, model versioning, and automated retraining.
• Integrated OpenTelemetry + Grafana for real-time monitoring of LLM drift, token usage, and response quality in production.
• Led code reviews, mentored 2 junior engineers, and collaborated with product managers to align AI solutions with HIPAA and enterprise compliance.

AI Engineer - Oriana Software Solutions - India

(2022-07 - 2023-12)

• Architected and maintained RESTful microservices using Spring Boot and Flask, handling 10k+ requests/minute for demand forecasting systems.
• Built ETL pipelines with Pandas, NumPy, and SQL to clean and transform 5M+ records for predictive modeling, reducing data prep time by 30%.
• Integrated Scikit-learn models into production for automated demand forecasting, improving forecast accuracy by 20% and reducing stockouts.
• Optimized SQL queries with indexing, partitioning, and connection pooling, cutting average query latency from 800ms to 250ms.
• Containerized applications with Docker and managed AWS EC2 deployments, achieving 99.9% uptime across staging and production environments.
• Actively participated in Agile/Scrum ceremonies, conducted peer code reviews, and maintained 85%+ test coverage using PyTest and JUnit.

AI Engineer Intern - Oriana Software Solutions - India

(2022-01 - 2022-06)

• Developed backend services and REST APIs for internal tools, enabling real-time access to customer analytics.
• Automated manual reporting workflows using Python scripts, reducing weekly effort by 40%.
• Wrote complex SQL queries and stored procedures for data extraction and business logic.
• Assisted in building customer segmentation models using Scikit-learn (K-Means, PCA).
• Participated in debugging, testing, and documentation to ensure production-ready code quality.

M.S. - Computer Science Management - Avila University