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

Full-Stack AI/ML Engineer with 8+ years of experience designing, building, and deploying production machine learning and generative AI systems across Banking, Life Sciences, and Asset Management. Experienced in agentic RAG and multi-agent orchestration using MCP/tool calling, along with LLM-powered assistants in regulated environments. Strong backend and REST API development background with hands-on work across AWS, GCP, and Azure, taking prototypes to reliable cloud-native production using Docker, Kubernetes, Terraform, and CI/CD. Skilled in semantic search (embeddings/vector DBs), LLM evaluation and hallucination detection, and production-grade AI guardrails, monitoring, and observability.…Full-Stack AI/ML Engineer with 8+ years of experience designing, building, and deploying production machine learning and generative AI systems across Banking, Life Sciences, and Asset Management. Experienced in agentic RAG and multi-agent orchestration using MCP/tool calling, along with LLM-powered assistants in regulated environments. Strong backend and REST API development background with hands-on work across AWS, GCP, and Azure, taking prototypes to reliable cloud-native production using Docker, Kubernetes, Terraform, and CI/CD. Skilled in semantic search (embeddings/vector DBs), LLM evaluation and hallucination detection, and production-grade AI guardrails, monitoring, and observability.

Prathyusha Reddy

Full Stack Developer, AI Engineer, Data Scientist, +4





Full-Stack AI/ML Engineer with 8+ years of experience designing, building, and deploying production machine learning and generative AI systems across Banking, Life Sciences, and Asset Management. Experienced in agentic RAG and multi-agent orchestration using MCP/tool calling, along with LLM-powered assistants in regulated environments. Strong backend and REST API development background with hands-on work across AWS, GCP, and Azure, taking prototypes to reliable cloud-native production using Docker, Kubernetes, Terraform, and CI/CD. Skilled in semantic search (embeddings/vector DBs), LLM evaluation and hallucination detection, and production-grade AI guardrails, monitoring, and observability.…Full-Stack AI/ML Engineer with 8+ years of experience designing, building, and deploying production machine learning and generative AI systems across Banking, Life Sciences, and Asset Management. Experienced in agentic RAG and multi-agent orchestration using MCP/tool calling, along with LLM-powered assistants in regulated environments. Strong backend and REST API development background with hands-on work across AWS, GCP, and Azure, taking prototypes to reliable cloud-native production using Docker, Kubernetes, Terraform, and CI/CD. Skilled in semantic search (embeddings/vector DBs), LLM evaluation and hallucination detection, and production-grade AI guardrails, monitoring, and observability.

Available to hire

Full-Stack AI/ML Engineer with 8+ years of experience designing, building, and deploying production machine learning and generative AI systems across Banking, Life Sciences, and Asset Management. Experienced in agentic RAG and multi-agent orchestration using MCP/tool calling, along with LLM-powered assistants in regulated environments.

Strong backend and REST API development background with hands-on work across AWS, GCP, and Azure, taking prototypes to reliable cloud-native production using Docker, Kubernetes, Terraform, and CI/CD. Skilled in semantic search (embeddings/vector DBs), LLM evaluation and hallucination detection, and production-grade AI guardrails, monitoring, and observability.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Beginner

Language

Work Experience

Senior Full-Stack AI /ML Engineer at Wells Fargo

May 1, 2024 - Present

Full-stack contributor on Fargo, a generative AI banking assistant used by tens of millions of retail customers, and an AI-driven compliance/risk monitoring platform used by financial-crime investigators. Developed and enhanced RAG components (ingestion, chunking, embeddings, metadata filtering, semantic retrieval, context assembly) and evolved retrieval toward agentic RAG with controlled tool/data-source selection. Integrated Pinecone for production semantic retrieval. Trained/tuned intent classification and NLU models using TensorFlow/PyTorch with SQL and BigQuery feature engineering. Deployed and monitored models via Google Vertex AI and containerized inference via Cloud Run to reduce inference cost while meeting quality targets. Built CI/CD and automated LLM evaluation pipelines using RAGAS for retrieval quality, faithfulness/hallucination detection, context precision/recall, and safety. Supported compliance-critical workflows with monitoring/observability and responsible AI contro

Senior Full-Stack AI/ML Engineer at Wells Fargo

May 1, 2024 - Present

Full-stack contributor to Fargo, a generative AI banking assistant serving tens of millions of retail customers, and an AI-driven compliance/risk monitoring platform used by financial-crime investigators. Built production RAG components including ingestion, chunking, embeddings, metadata filtering, semantic retrieval, and context assembly. Helped evolve retrieval toward agentic RAG with dynamic tool/data-source selection via LangChain/LlamaIndex and later MCP-based integrations for controlled access. Integrated Pinecone for millisecond-to-low-second semantic retrieval. Developed and improved intent classification/NLU using TensorFlow/PyTorch with SQL/BigQuery feature engineering. Deployed and monitored models using Vertex AI and containerized inference via Cloud Run, reducing inference cost ~30–40% while meeting quality targets. Built CI/CD and automated LLM evaluation pipelines (RAGAS) for retrieval quality, faithfulness/hallucination detection, and safety, improving responsible-AI

Full-Stack AI /ML Engineer at Takeda

April 1, 2020 - November 30, 2023

Modernized genomics pipeline by migrating on-prem RNA sequencing analysis workflows to AWS HealthOmics, reducing turnaround time from six weeks to two days and costs by ~70% across 20,000+ samples. Built supporting AWS infrastructure (S3, IAM, AWS Batch/HealthOmics) and internal tooling/dashboards using React and Python for researchers to submit samples, monitor runs, and review results. Implemented CI/CD automation with Git and CloudWatch for logging/metrics. Built myAibou, a production-oriented internal generative AI assistant for company-wide employee productivity without exposing data to public AI systems. Migrated to AWS Bedrock, implemented Python backend services on AWS Lambda with API Gateway, and built a React chat UI hosted on AWS Amplify. Created a RAG pipeline using OpenSearch vector search and Titan Embeddings for meeting transcript search, and implemented tool/function calling for actions like pulling calendar details or summarizing documents. Added prompt versioning/tes

Full-Stack AI/ML Engineer at Takeda

April 1, 2020 - November 30, 2023

Modernized genomics analysis workflows by migrating on-prem genomic pipelines to AWS HealthOmics, processing 20,000+ RNA-seq samples; reduced turnaround time from six weeks to two days and reduced costs by ~70%. Supported pipeline infrastructure using S3, IAM, AWS Batch, and HealthOmics. Built internal tooling and dashboards using React and Python for researchers to submit samples, monitor runs, and review results. Implemented CI/CD with Git/GitHub Actions and monitored via CloudWatch. Built myAibou, a production-oriented internal generative AI assistant using OpenAI API and later AWS Bedrock. Implemented Python backend services on AWS Lambda with REST APIs via API Gateway, hosted a React chat UI with AWS Amplify, and developed a RAG pipeline using OpenSearch vector storage and Titan embeddings for meeting transcript search. Added function/tool calling (e.g., calendar pulls, document summarization) and established prompt versioning and regression testing with CI/CD and CloudWatch.

Python AI Developer at Vanguard

August 1, 2017 - April 30, 2020

Worked within Vanguard’s Quantitative Equity Group and AI Garage on data pipelines, ensemble model development, and validation for systematic active equity strategies. Built Python/SQL/PySpark pipelines to extract, cleanse, and transform large-scale datasets from Snowflake and enterprise sources. Developed reusable feature engineering modules covering macroeconomic, valuation, and price/volume signals. Implemented and maintained an ensemble modeling framework using Scikit-learn and Statsmodels. Produced regression/classification/ensemble model code and backtesting infrastructure to validate models across historical periods and simulated portfolios. Built configurable tuning/feature-selection scripts, reducing testing time from days to hours.