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

AI/ML Engineer with ~3 years of experience building production-scale Generative AI, LLM, and machine learning systems using Python and PyTorch. Strong focus on RAG and AI agent orchestration with LangChain/LangGraph, evaluation-driven improvements, and scalable retrieval pipelines. Experience delivering intelligent document processing and multimodal document intelligence using AWS (Textract/Bedrock), vector search, and robust MLOps (MLflow, Kubernetes, Docker) to reduce latency and improve accuracy. Skilled in integrating LLM solutions into secure, cloud-native enterprise applications.…AI/ML Engineer with ~3 years of experience building production-scale Generative AI, LLM, and machine learning systems using Python and PyTorch. Strong focus on RAG and AI agent orchestration with LangChain/LangGraph, evaluation-driven improvements, and scalable retrieval pipelines. Experience delivering intelligent document processing and multimodal document intelligence using AWS (Textract/Bedrock), vector search, and robust MLOps (MLflow, Kubernetes, Docker) to reduce latency and improve accuracy. Skilled in integrating LLM solutions into secure, cloud-native enterprise applications.

Hrishikesh Rajaputra

AI Engineer, Data Scientist, Software Engineer, +2





AI/ML Engineer with ~3 years of experience building production-scale Generative AI, LLM, and machine learning systems using Python and PyTorch. Strong focus on RAG and AI agent orchestration with LangChain/LangGraph, evaluation-driven improvements, and scalable retrieval pipelines. Experience delivering intelligent document processing and multimodal document intelligence using AWS (Textract/Bedrock), vector search, and robust MLOps (MLflow, Kubernetes, Docker) to reduce latency and improve accuracy. Skilled in integrating LLM solutions into secure, cloud-native enterprise applications.…AI/ML Engineer with ~3 years of experience building production-scale Generative AI, LLM, and machine learning systems using Python and PyTorch. Strong focus on RAG and AI agent orchestration with LangChain/LangGraph, evaluation-driven improvements, and scalable retrieval pipelines. Experience delivering intelligent document processing and multimodal document intelligence using AWS (Textract/Bedrock), vector search, and robust MLOps (MLflow, Kubernetes, Docker) to reduce latency and improve accuracy. Skilled in integrating LLM solutions into secure, cloud-native enterprise applications.

Available to hire

AI/ML Engineer with ~3 years of experience building production-scale Generative AI, LLM, and machine learning systems using Python and PyTorch. Strong focus on RAG and AI agent orchestration with LangChain/LangGraph, evaluation-driven improvements, and scalable retrieval pipelines.

Experience delivering intelligent document processing and multimodal document intelligence using AWS (Textract/Bedrock), vector search, and robust MLOps (MLflow, Kubernetes, Docker) to reduce latency and improve accuracy. Skilled in integrating LLM solutions into secure, cloud-native enterprise applications.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Work Experience

AI/ML Engineer at Citi

July 1, 2025 - Present

Built a production-grade RAG platform using LangChain, LangGraph, Azure OpenAI, and vector databases to automate financial knowledge discovery across regulatory documents, reducing analyst research effort by ~45%. Engineered scalable document intelligence pipelines using Python, PyTorch, and Hugging Face Transformers to process 400K+ financial documents monthly and improve semantic retrieval quality. Designed LLM orchestration workflows with LangGraph and prompt optimization across Claude/GPT models, reducing irrelevant responses by ~38% via evaluation-driven tuning. Implemented MLOps workflows with MLflow and Kubernetes for automated evaluation, versioning, deployment, and monitoring—cutting deployment time from 3 days to under 12 hours. Optimized vector search using FAISS/Pinecone, metadata filtering, and hybrid retrieval to reduce latency from ~2.5s to under 900ms. Added human-in-the-loop validation and monitoring dashboards, and integrated GenAI services into financial platforms

AI / ML Engineer at Accenture

September 1, 2023 - June 30, 2025

Architected a production-grade intelligent document processing (IDP) platform using Python, TensorFlow, OpenCV, AWS Textract, and FastAPI to automate document classification, extraction, and validation across 8M+ documents annually, reducing manual review effort. Developed multimodal document intelligence models using TensorFlow, Amazon Bedrock, LangChain, and NLP with 97% extraction accuracy for classification, entity extraction, and contextual summarization. Built high-throughput computer vision pipelines using OpenCV/TensorFlow/OCR for low-quality scanned documents, improving OCR accuracy by ~24%. Implemented production RAG pipelines using Amazon Bedrock, LangChain, and PostgreSQL for citation-grounded search and compliance validation, improving retrieval relevance by ~31%. Operationalized end-to-end MLOps with MLflow, AWS, and CI/CD for faster deployments and improved release reliability. Created scalable data pipelines (Python/SQL/Pandas/PostgreSQL) reducing pipeline execution tim