Open to opportunities

Omeir Ash

@omeirash

Message

Senior AI/ML Engineer building production GenAI, RAG, agentic AI, and MLOps systems.

United States

Message

What I'm looking for

I’m looking for a team where I can ship governed GenAI/LLM systems to production—designing RAG and agentic AI workflows, optimizing latency and cost, and partnering cross-functionally to deliver measurable, reliable outcomes.

I’m a Senior AI/ML Engineer with 8+ years building production machine learning and GenAI systems—especially LLM applications, RAG platforms, agentic AI workflows, and low-latency inference. I focus on moving AI from prototype to production while balancing accuracy, latency, cost, reliability, security, and user experience.

In my recent roles, I’ve improved documentation efficiency by 40%, increased retrieval relevance and grounding by 30%, reduced inference costs by 30%, processed 50M+ records, and shortened ML release cycles by 40%. I partner closely with product, platform, data engineering, risk, and business teams to translate complex requirements into measurable, governed AI systems with strong monitoring, evaluation, and auditability.

Experience

Work history, roles, and key accomplishments

Current

Senior AI Engineer

Current

Abridge

Dec 2023 - Present (2 years 6 months)

Architected production LLM pipelines using OpenAI/Anthropic Claude and Hugging Face, improving documentation efficiency by 40%. Built RAG and agentic AI workflows, increasing retrieval relevance by 30% and task completion reliability by 28%, and launched low-latency inference to cut serving costs by 30%.

Generative AI LLM RAG Agentic AI Hybrid Retrieval & Reranking LLM Inference TorchServe Kubernetes Real Time Data Pipelines (Kafka Airflow)

Machine Learning Engineer

Zest AI

Jun 2020 - Dec 2023 (3 years 6 months)

Delivered production predictive models with XGBoost/LightGBM and deep learning, improving accuracy by 20% for decisioning workflows. Processed 50M+ records and automated SageMaker CI/CD and MLflow governance to shorten ML release cycles by 40%, while reducing real-time inference latency by 30%.

XGBoost LightGBM scikit learn Deep Learning CNN AWS Sagemaker MLFlow LIME

Associate AI/ML Engineer

Netomi

Jul 2017 - May 2020 (2 years 10 months)

Built conversational AI systems using NLP/transformer models to increase automation coverage by 30%. Improved semantic search and retrieval accuracy by 25% and strengthened intent classification performance by 20% through end-to-end NLP pipelines and production AWS deployments.

NLP Conversational AI Transformers Intent Classification & Dialogue Management