Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI/ML Engineer with 4 years of experience building GenAI applications, machine learning pipelines, backend services, and cloud-based model APIs across banking, media, and financial services. Skilled in Python, SQL, LangChain, Azure OpenAI, Hugging Face, PyTorch, FastAPI, and production MLOps. Experienced in RAG workflows, secure API development, speech-to-text pipelines, recommendations, feature engineering, data validation, model monitoring, and production AI/ML deployment. Proven impact includes reducing review time for captioning workflows, improving RAG answer relevance, and building scalable model services with Docker, Kubernetes, and managed cloud platforms.…AI/ML Engineer with 4 years of experience building GenAI applications, machine learning pipelines, backend services, and cloud-based model APIs across banking, media, and financial services. Skilled in Python, SQL, LangChain, Azure OpenAI, Hugging Face, PyTorch, FastAPI, and production MLOps. Experienced in RAG workflows, secure API development, speech-to-text pipelines, recommendations, feature engineering, data validation, model monitoring, and production AI/ML deployment. Proven impact includes reducing review time for captioning workflows, improving RAG answer relevance, and building scalable model services with Docker, Kubernetes, and managed cloud platforms.

Durga Sai Saran

AI Engineer, Software Engineer, Data Scientist, +2





AI/ML Engineer with 4 years of experience building GenAI applications, machine learning pipelines, backend services, and cloud-based model APIs across banking, media, and financial services. Skilled in Python, SQL, LangChain, Azure OpenAI, Hugging Face, PyTorch, FastAPI, and production MLOps. Experienced in RAG workflows, secure API development, speech-to-text pipelines, recommendations, feature engineering, data validation, model monitoring, and production AI/ML deployment. Proven impact includes reducing review time for captioning workflows, improving RAG answer relevance, and building scalable model services with Docker, Kubernetes, and managed cloud platforms.…AI/ML Engineer with 4 years of experience building GenAI applications, machine learning pipelines, backend services, and cloud-based model APIs across banking, media, and financial services. Skilled in Python, SQL, LangChain, Azure OpenAI, Hugging Face, PyTorch, FastAPI, and production MLOps. Experienced in RAG workflows, secure API development, speech-to-text pipelines, recommendations, feature engineering, data validation, model monitoring, and production AI/ML deployment. Proven impact includes reducing review time for captioning workflows, improving RAG answer relevance, and building scalable model services with Docker, Kubernetes, and managed cloud platforms.

Available to hire

AI/ML Engineer with 4 years of experience building GenAI applications, machine learning pipelines, backend services, and cloud-based model APIs across banking, media, and financial services. Skilled in Python, SQL, LangChain, Azure OpenAI, Hugging Face, PyTorch, FastAPI, and production MLOps.

Experienced in RAG workflows, secure API development, speech-to-text pipelines, recommendations, feature engineering, data validation, model monitoring, and production AI/ML deployment. Proven impact includes reducing review time for captioning workflows, improving RAG answer relevance, and building scalable model services with Docker, Kubernetes, and managed cloud platforms.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Beginner

Beginner

Beginner

Language

Work Experience

AI/ML Engineer at Warner Bros. Discovery

September 1, 2025 - Present

Engineered Python-based speech-to-text pipelines using Vertex AI and Cloud Storage to support caption automation, shortening review preparation time by 35% for unscripted content. Trained transformer models with PyTorch and Hugging Face on cleaned media transcripts to improve caption accuracy, reducing repeated manual corrections by 18%. Deployed inference services with Docker, Kubernetes, and FastAPI to expose caption scoring APIs. Optimized BigQuery/SQL feature tables for viewer behavior analysis and created recommendation prototypes with embeddings and vector search to increase click-through rate by 12%. Implemented model drift/latency monitoring using MLflow and CloudWatch, and used SHAP/error sampling for QA explainability. Streamlined Airflow transcript ingestion and batch inference pipelines, and validated LLM/pipeline outputs against human-reviewed samples for accessibility workflows.

Software Engineer - Generative AI at M&T Bank

July 1, 2024 - August 31, 2025

Implemented RAG workflows using Python, LangChain, Azure OpenAI, and FAISS to answer banking policy questions, cutting support research time by 32%. Built secure FastAPI services connecting LLM outputs with internal knowledge sources. Improved prompt templates and retrieval filters to reduce unsupported responses and increase answer relevance by 24% in controlled QA. Orchestrated Airflow document ingestion, chunking, embedding generation, and metadata tagging to keep support content current. Secured model API traffic with authentication checks, logging, and masking rules to protect sensitive banking data. Tuned Hugging Face embedding models and vector search parameters to improve retrieval for fraud/digital banking/branch support queries. Containerized GenAI services with Docker and Kubernetes deployment artifacts to reduce environment setup issues by 27%. Reviewed MLflow metrics and defects with product owners to guide safer banking AI feature backlog priorities.

Software Engineer at Hexaware Technologies

June 1, 2020 - August 31, 2022

Developed Java Spring Boot REST APIs for financial workflow modules to streamline transaction processing and reduce manual validation effort by 28%. Built backend services using Java, SQL, and Hibernate for customer onboarding and account management. Analyzed legacy logic and optimized SQL queries to address performance bottlenecks, improving API response times by 22%. Integrated role-based access controls and business validation rules for security and regulatory compliance. Designed SQL-based extraction/reporting processes to consolidate operational data, improving accuracy and reducing manual reporting effort by 30%. Evaluated logs/transaction trends to identify recurring incidents and improve high-priority incident resolution by 35% per sprint. Collaborated with cross-functional teams in Agile delivery to stabilize enhancements and improve unit test coverage from 62% to 78%. Prepared datasets and validated business rules to support analytics and future data-driven/AI solution develo