Min Oh Email & Phone Number

Kirkland, Wa, Us

Redmond, Washington, United States

Tech lead at Research, Science & Strategy team- Generative AI, LLM, Prompt optimization, Hallucination evaluation and mitigation, Multi-modal learning- Multi-omics, Drug discovery, Chemotherapy response, Disease diagnostics

Redmond, Washington, United States

• Led multiple research initiatives in building reference-based Science Copilot, evaluating LLM hallucinations, and automatic prompt optimization• Led a cross-company science team of 9+ scientists and was responsible for executing and coordinating research efforts in omics and clinical data-driven disease outcome prediction• Developed fundamental AI/ML models for drug development process

Redmond, Washington, United States

• Developed a language model localizing culprit commits causing network incidents• Proposed new KPIs for service quality that brought new insights into leadership across organizations• Implemented dashboards and data processing pipelines for analyzing system telemetry and quality metrics

Blacksburg, Virginia

• Deep Learning for Enhancing Precision Medicine

Blacksburg, Virginia

• Developed a computational pipeline for prioritizing environmental resistome risk based on metagenomics data • Devised a quantitative risk measure based on the detection of antibiotic resistance genes, mobile genetic elements, human pathogens in environmental metagenomics data • Applied best practices for quality control, assembly, annotation of microbiome data

Blacksburg, Virginia

• CS 5824: Advanced Machine Learning • CS 3654: Introductory Data Analytics and Visualization • CS 1044: Introduction to Programming in C++

• Developed a hybrid ML-model for predicting CPU utilization on AFD nodes based on traffic data.• Implemented an end-to-end ML service that automatically trains, tests, monitors, and deploys the ML model on the configured schedule.• Transplanted ML model into a traffic simulation platform in production and improved confidence in risk assessment.

• Developed a data-driven triage bot that automatically routes network incidents based on end-to-end logs • Deployed the triage bot in production and it outperformed human in triage accuracy and saved hundreds of man-hours within a month • Designed and implemented an hop-analysis pipeline that enabled new understandings of incidents

Greater Seattle Area

• Developed unsupervised deep learning models for next-generation sequencing data • Implemented deep embedding techniques for representation learning of variant calling data • Successfully reconstructed population structure of different origins based on the high-dimensional and low sample size genomic data

Greater Seattle Area

• Developed a machine learning R package importing and analyzing VCF files in R environment• Implemented regression models (e.g. Neural Net) for estimating the quality of variant calls• Analyzed annotation modules of variant calling pipeline (BWA+GATK) on the Azure cloud

Seongnam, Gyeonggi-Do, Korea

• Designed a computational platform for deriving drug-specific pathways at a molecular level• Developed network-based classification models for predicting drug repositioning candidates and potential side effects• Implemented machine learning models for predicting breast cancer prognosis based on gene expression and protein-protein interaction data

Doctor Of Philosophy (Ph.D.), Computer Science

Bachelor Of Engineering (B.E.), Computer Engineering

Military Service 2008-2010