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

AI engineer with 4+ years of experience building Generative AI, RAG, and machine learning solutions for business use cases. Experienced with LLMs (GPT, Llama, Claude), NLP/deep learning, and agentic AI workflows using LangChain and LangGraph. Strong end-to-end delivery across data processing/feature engineering, scalable API and microservices development, and production MLOps (CI/CD, monitoring, drift detection, and retraining). Skilled in cloud-native deployments on AWS/Azure/GCP with robust governance, security, and evaluation to improve reliability, reduce latency, and optimize cost.…AI engineer with 4+ years of experience building Generative AI, RAG, and machine learning solutions for business use cases. Experienced with LLMs (GPT, Llama, Claude), NLP/deep learning, and agentic AI workflows using LangChain and LangGraph. Strong end-to-end delivery across data processing/feature engineering, scalable API and microservices development, and production MLOps (CI/CD, monitoring, drift detection, and retraining). Skilled in cloud-native deployments on AWS/Azure/GCP with robust governance, security, and evaluation to improve reliability, reduce latency, and optimize cost.

Prathyush LAI

AI Engineer, AI Developer, Data Scientist, +3





AI engineer with 4+ years of experience building Generative AI, RAG, and machine learning solutions for business use cases. Experienced with LLMs (GPT, Llama, Claude), NLP/deep learning, and agentic AI workflows using LangChain and LangGraph. Strong end-to-end delivery across data processing/feature engineering, scalable API and microservices development, and production MLOps (CI/CD, monitoring, drift detection, and retraining). Skilled in cloud-native deployments on AWS/Azure/GCP with robust governance, security, and evaluation to improve reliability, reduce latency, and optimize cost.…AI engineer with 4+ years of experience building Generative AI, RAG, and machine learning solutions for business use cases. Experienced with LLMs (GPT, Llama, Claude), NLP/deep learning, and agentic AI workflows using LangChain and LangGraph. Strong end-to-end delivery across data processing/feature engineering, scalable API and microservices development, and production MLOps (CI/CD, monitoring, drift detection, and retraining). Skilled in cloud-native deployments on AWS/Azure/GCP with robust governance, security, and evaluation to improve reliability, reduce latency, and optimize cost.

Available to hire

AI engineer with 4+ years of experience building Generative AI, RAG, and machine learning solutions for business use cases. Experienced with LLMs (GPT, Llama, Claude), NLP/deep learning, and agentic AI workflows using LangChain and LangGraph.

Strong end-to-end delivery across data processing/feature engineering, scalable API and microservices development, and production MLOps (CI/CD, monitoring, drift detection, and retraining). Skilled in cloud-native deployments on AWS/Azure/GCP with robust governance, security, and evaluation to improve reliability, reduce latency, and optimize cost.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Language

English

Fluent

Work Experience

AI Engineer at Wizards of the Coast

August 1, 2025 - Present

Designed and implemented end-to-end Retrieval-Augmented Generation (RAG) architectures, integrating vector databases, embedding pipelines, and LLM orchestration frameworks to enable context-aware and high-precision responses. Built scalable document intelligence pipelines for PDF parsing, OCR-based text extraction, embedding generation, and semantic search to process large volumes of unstructured enterprise data. Developed and optimized PyTorch models for classification, recommendation, and user engagement, improving predictive performance and system scalability. Reduced RAG re-indexing latency by 60% through efficient chunking, asynchronous embeddings, and streamlined vector storage. Deployed low-latency inference services via FastAPI-based microservices with monitoring and automated retraining. Built agentic AI systems using LangChain and LangGraph with tool-calling and multi-step reasoning. Implemented LLM workflows with context management, guardrails, and robust access controls to

AI/ML Engineer at Verisk Analytics

August 1, 2024 - July 1, 2025

Optimized Snowflake data models and complex SQL transformations to support scalable AI/ML workloads and Retrieval-Augmented Generation (RAG). Designed and implemented end-to-end RAG architectures with embedding generation, vector similarity search, semantic retrieval, and context injection to enable accurate, context-aware LLM responses. Built high-performance PyTorch inference services integrated with backend systems and distributed pipelines, supporting real-time and batch processing with low latency. Developed NLP and predictive analytics models to enhance claims processing and underwriting workflows, leading to better decision-making. Engineered secure AWS-based infrastructure (VPC, IAM, encryption) and robust data ingestion pipelines for structured and unstructured data (contracts, claims, policy documents). Improved claims processing efficiency by 35% via multi-agent LLM-based summarization, automated retrieval, and decision-support. Fine-tuned transformer models on domain-specif

Data Scientist at eSparkBiz Technologies

June 1, 2021 - July 1, 2023

Implemented robust data ingestion frameworks using Python and Azure Data Factory to consolidate financial statements, macroeconomic indicators, and historical rating data from diverse sources. Designed scalable feature engineering pipelines using Pandas, NumPy, and PySpark for risk modeling and temporal analysis. Built and deployed predictive ML solutions (classification, regression, clustering) to improve credit risk assessment, rating outlooks, and issuer performance tracking. Developed deep learning models (LSTM, GRU, CNN) to extract insights from financial narratives and disclosures. Productionized ML models as scalable RESTful services via FastAPI and Flask on AKS, ensuring low latency and high availability. Implemented SHAP/LIME-based interpretability and built end-to-end MLOps with Azure ML, MLflow, and Azure DevOps. Built data ingestion pipelines for structured/unstructured data (contracts, claims, policies) enabling downstream AI/ML. Achieved ~18% improvement in predictive acc