Wilmer E.

Key Skills

Software

No software listed

Top Subject Matter

STEM (materials science/chemistry/physics)

sustainable materials ranking

intermetallic coatings

Top Data Types

Computer Code Programming

Top Task Types

RLHF

Freelancer Overview

AI model/rubric evaluation for STEM candidate ranking via Natural Occurrence Score (NOS) framework. Brings 23+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include GitHub and Python. AI-training focus includes data types such as Computer Code and Programming and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

Scientific workflow review and code contributions supporting research evaluation (ComProScanner)

Reviewed and improved a scientific software workflow (ComProScanner) intended for Scopus queries and research-platform operations. Contributed to code review and issue/PR handling to make the workflow more reliable for downstream information retrieval and validation. Supported structured workflow improvements that function as a form of labeled/curated query processing for research evaluation. • Installed and code-reviewed ComProScanner workflow • Contributed issues/PRs to refine query/processing behavior • Improved reproducible scientific software workflows for STEM evaluation • Supported validation-oriented workflow design for research queries

2026 - Present

RLHF-style STEM model evaluation and rubric-based review for manufacturing guidance

RLHF

Conducted RLHF/model evaluation work in STEM and industrial contexts to assess model outputs for scientific and manufacturing relevance. Designed evaluation approaches that consider whether generated manufacturing instructions are realistic, safe, complete, and technically coherent. Combined industrial reasoning with STEM-focused evaluation criteria to improve trustworthiness of model responses. • RLHF or related evaluation/feedback loops for STEM and manufacturing instruction quality • Safety, completeness, and technical coherence checks on model-generated instructions • Rubric-based rating of outputs for evaluation purposes • Integration of domain constraints into the assessment workflow

2026 - Present

AI model/rubric evaluation for STEM candidate ranking via Natural Occurrence Score (NOS) framework

Developed and applied the Natural Occurrence Score (NOS) framework to rank intermetallic coating candidates using sustainability, scalability, thermodynamic stability, and mechanical descriptors. Performed rubric-based evaluation and confidence/error analysis workflows intended to support STEM model evaluation and candidate screening. Used Python and reproducible notebooks to compute scores and validate candidate rankings against domain expectations. • STEM model evaluation and rubric/workflow design for ranking and validation • Sustainability screening using abundance/criticality and supply-risk style descriptors • Candidate ranking with stability and mechanical descriptor inputs • Error analysis and validation strategy for confidence scoring

2026 - Present

Education

Work History