Portrait of Joseph M. Weaver

Biographical Sketch

Joseph M. Weaver

Ph.D. student in Statistics at Michigan State University focused on Bayesian crop-risk modeling, intrinsic dimension learning, and large-scale geospatial and HPC pipelines.

weave151@msu.edu LinkedIn

About

Consultant with Land Core building Bayesian crop-yield risk models and geospatial data systems (5+ TB), and senior AI engineer background leading distributed data and ML platforms at Jackson National Life. I blend statistical theory, reproducible pipelines, and scalable infrastructure across MSU HPCC, Spark, PostGIS, DuckDB, and containerized workflows.

Highlights

  • Designed ETL and risk modeling pipelines for multi-terabyte geospatial datasets in PostGIS and DuckDB.
  • Built containerized Spark and HPC workflows with CI/CD, self-healing, and compliance-first design.
  • Instructor for STT 200; mentor bridging statistical learning, stochastic processes, and applied computing.

Research & Interests

  • Bayesian hierarchical crop yield risk modeling and uncertainty quantification.
  • Intrinsic dimension estimation and manifold learning for high-dimensional data.
  • Reproducible research pipelines: YAML metadata, GitHub Actions, and automated documentation.

Current Roles

  • Ph.D. Student, Statistics & Probability, Michigan State University (in progress)
  • Consultant, Land Core - data pipelines and crop-risk modeling (2022-present)
  • Graduate Instructor (TA), STT 200: Introduction to Statistics

Selected Prior Experience

  • Senior AI Engineer, Jackson National Life (2018-2022): led AI/data engineering architecture and Spark pipelines.
  • Senior Capacity Planner, Jackson National Life (2011-2018): automated regression/cluster forecasting for infra.
  • Senior Software Developer, Jackson National Life (2005-2011): unified .NET automation for nationwide broker-dealer systems.

Collaboration

Collaborating with teams at Rice University and Land Core on Bayesian crop-risk modeling, geospatial ETL, and reproducible workflows that connect theory to decision systems. Open to research collaborations and data engineering engagements.

Comments