Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am an analytically driven computer scientist with an MS from The University of Texas at Austin and 6+ years of experience in AI data quality assurance, LLM output evaluation, and machine learning model auditing. I apply structured rubrics to assess AI-generated content for factual accuracy, hallucinations, and alignment with user intent, and I bring a passion for pop culture, streaming media, and entertainment to evaluating model responses across cinematic universes, episodic content, and digital media.…I am an analytically driven computer scientist with an MS from The University of Texas at Austin and 6+ years of experience in AI data quality assurance, LLM output evaluation, and machine learning model auditing. I apply structured rubrics to assess AI-generated content for factual accuracy, hallucinations, and alignment with user intent, and I bring a passion for pop culture, streaming media, and entertainment to evaluating model responses across cinematic universes, episodic content, and digital media.

Sheila Elvira Motoni

Data Scientist, Data Analyst, AI Engineer, +2





I am an analytically driven computer scientist with an MS from The University of Texas at Austin and 6+ years of experience in AI data quality assurance, LLM output evaluation, and machine learning model auditing. I apply structured rubrics to assess AI-generated content for factual accuracy, hallucinations, and alignment with user intent, and I bring a passion for pop culture, streaming media, and entertainment to evaluating model responses across cinematic universes, episodic content, and digital media.…I am an analytically driven computer scientist with an MS from The University of Texas at Austin and 6+ years of experience in AI data quality assurance, LLM output evaluation, and machine learning model auditing. I apply structured rubrics to assess AI-generated content for factual accuracy, hallucinations, and alignment with user intent, and I bring a passion for pop culture, streaming media, and entertainment to evaluating model responses across cinematic universes, episodic content, and digital media.

Available to hire

I am an analytically driven computer scientist with an MS from The University of Texas at Austin and 6+ years of experience in AI data quality assurance, LLM output evaluation, and machine learning model auditing.

I apply structured rubrics to assess AI-generated content for factual accuracy, hallucinations, and alignment with user intent, and I bring a passion for pop culture, streaming media, and entertainment to evaluating model responses across cinematic universes, episodic content, and digital media.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

AI Collection (Audio)

Intermediate

AI Collection (Video)

Intermediate

Bash

Intermediate

GitHub Copilot

Intermediate

Work Experience

Senior AI Data Quality Analyst at Oracle Corporation

February 1, 2022 - Present

Lead quality audits of AI-generated outputs for Oracle's cloud-based customer service LLM, reviewing 300+ model responses weekly against a multi-dimensional rubric covering accuracy, coherence, and user query fulfillment. Identify hallucinations, factual errors, and missing information in model-generated content across technology, general knowledge, and entertainment domains; produce detailed error taxonomy reports for model fine-tuning teams. Evaluate contributor annotation quality for RLHF datasets, verifying reviewers apply rubric dimensions and justification criteria before data enters training pipelines. Design and implement standardized output-scoring rubric adopted across a 40-person evaluation team, improving inter-annotator agreement scores by 21%. Conduct red-team exercises on LLM endpoints, probing adversarial responses, hallucinations, and policy violations in entertainment, news, and general knowledge domains.

Data Analyst II at H-E-B Digital

July 1, 2019 - January 1, 2022

Analyzed customer search query datasets and evaluated AI-generated product recommendations against curated ground-truth labels, applying structured relevance-scoring rubrics. Built Python-based data validation pipelines to detect annotation inconsistencies across 1M+ labeled records, reducing label error rate by 17%. Authored weekly quality assessment reports for leadership, highlighting model performance gaps and recommending rubric refinements. Collaborated with NLP engineers to review and refine training data for H-E-B's search ranking model, focusing on user intent alignment across grocery, general merch and digital media domains.