Machine Learning Engineer

Nebulla One December****-February****

Machine Learning Engineer

Developed and implemented time series models (ARIMA, Prophet, LSTM) in Dataiku with 45% enhancement in forecast accuracy in revenue and inventory.

Implemented ML workflows in Dataiku DSS, reducing model deployment time by 40% with Python and SQL.

Implemented LLM-based tools for search and financial summarization, reducing review time for analysts by 70%.

Developed ML models (XGBoost, logistic regression, clustering) for risk and churn prediction that enhanced retention by 25%.

Created A/B testing workflows in Dataiku for model and campaign testing with statistical integrity.

Mentored 4 junior data scientists, improving team efficiency by 20%.

Cross-functionally collaborated to turn data into insights that yielded $1.5M in operational savings.

Excell Valley November2018-August2021

Machine Learning Engineer

Implemented pipelines ingest PDFs,Word docs, and HTML pages from S3, preprocess content,chunk text using Lang Chain, and generate Vector Embedding stored in FAISS

Designed a RAG System using embedding-based search, integrated with AWS Bedrock LLMs to answer questions integrating it with Fast API and AWS API Gateway.

Implemented a agent planner for handling multi-step queries mainly document retrieval and summarization

Fine Tuned prompts using prompt-chain with AWS Bedrock to generate responses.

Worked on model metrics to assess answer accuracy integrating with retrieval logic and LLM prompt strategies

Containerized the solution with Docker, deployed to AWS EKS, and integrated with GitHub Actions Code Pipeline for automated testing and deployments

Monitored alerts and triggers using AWS Cloud Watch

Shelloris April 2014 - November 2018

Machine Learning Engineer

Implemented a RAG Pipeline using embedding models OpenAI,Hugging Face,Vector databases FAISS,Pinecone,and custom search ranking algorithms for semantic context retrieval of customer profiles.

Worked with Restful API’S using FastAPI and Flask, integrated with AWS API Gateway and event driven architecture using AWS Lambda, Step Functions, and CloudWatch for monitoring and orchestration

Built RAG pipelines using embedding models and vector databases for improved context retrieval and ranking.

Implemented LLM prompts for summarization,question answering,and classification tasks, achieving a 65% improvement in response

Built CI/CD pipelines for GenAI services using Docker, GitHub Actions, and AWS CodePipeline with EKS clusters

Executed agentic workflows using tools like Lang Chain and orchestration frameworks for customer support

Managed storage infrastructure using AWS S3, RDS, DynamoDB, and EKS

Collaborated with cross functional teams to build GENAI use case.

Contact this candidate