Avinash Yagnapu

Postdoctoral Researcher | Machine Learning for Scientific Discovery

Talent

Las VegasMember since July 28, 2026

PyTorchTensorFlowTransformers and Attention ArchitecturesSelf-Supervised and Contrastive Representation LearningWeakly Supervised and Positive-Unlabeled LearningReinforcement LearningInterpretabilityImaging (medical, pathology, microscopy)High-dimensional omics and genomicsStructured scientific and clinical dataComputer vision (OpenCV)Computational experiment designBenchmarkingModel performance robustness and generalizability evaluationInterpretability analysisPythonC#JavaJavaScriptBash/ShellUnix/LinuxCloud (AWS)Apache SparkDockerGit/GitHubReproducible research software

Hire this person

Send a job offer directly to this candidate

PyTorchTensorFlowTransformers and Attention ArchitecturesSelf-Supervised and Contrastive Representation LearningWeakly Supervised and Positive-Unlabeled Learning

About

Machine learning researcher (Ph.D., Computer Science, 2025) with 6+ years developing novel ML models and adapting state-of-the-art approaches for scientific applications. Strong theoretical and practical foundation in deep learning, including transformer and attention architectures, self-supervised and contrastive representation learning, and learning from incomplete or weakly labeled data. Experienced in multi-modality data analysis spanning imaging, structured scientific, and high-dimensional omics data, with excellent programming skills in Python, C++, and PyTorch.

Demonstrated publication record in machine learning and applied science venues, formulating research directions and communicating results within interdisciplinary collaborations alongside domain scientists.

Experience

Postdoctoral Associate at University of Nevada, Las Vegas (2026-01 – Present)

Develop novel machine learning models for scientific discovery, including transformer-based architectures and self-supervised contrastive representation learning, with rigorous evaluation of model performance and generalizability.
Formulate research directions independently and with mentors, and communicate progress and results through manuscripts, presentations, and interdisciplinary collaboration with domain scientists.

Graduate Research Assistant, DataXLab at University of Nevada, Las Vegas (2018-08 – 2025-12)

Developed and adapted deep learning models for scientific data across multiple modalities, including imaging, genomics, proteomics, multi-omics, and structured records, producing multiple peer-reviewed publications.
Designed novel architectures for learning under limited supervision, including multi-layered self-attention for weakly supervised segmentation (CVIU) and positive-unlabeled learning for molecular interaction prediction (Bioinformatics).
Built scalable, reproducible research software in Python and C/C++ within Unix/Linux, GPU, cloud (AWS), and distributed (Apache Spark) computing environments, working alongside subject matter experts.

Instructor, CS 138: Programming for Data Science in Python at University of Nevada, Las Vegas (2023-08 – 2023-12)

Designed and delivered course content, managed materials via Git, and earned an instructor evaluation score of 4.38 (Fall 2023).

Machine Learning Lab Instructor — UNLV International Student Program at University of Nevada, Las Vegas (2025-07 – 2025-09)

Mentored student cohorts through hands-on projects spanning single/multi-omics survival analysis and histopathology-based cancer classification, from data preparation to evaluation.