Jennifer B.

Key Skills

Software

Mercor

SuperAnnotate

Clickworker

Data Annotation Tech

Other

Top Subject Matter

LLM evaluation

safety alignment

and AI reliability

Top Data Types

Text

Audio

Image

Document

Top Task Types

RLHF

Transcription

Red Teaming

Freelancer Overview

Independent Contractor Software Developer/AI Trainer (LLM output validation, red teaming, evaluation). Brings 23+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Bachelor of Science, Iowa State University (2003). AI-training focus includes data types such as Text, Audio, and Image and labeling workflows including Red Teaming, Transcription, and Evaluation.

Labeling Experience

Software Developer/AI Trainer — Multimodal image evaluation with A/B testing and rubric rankings

Image

Conducted multimodal evaluation and A/B testing across computer vision pipelines to rank AI-generated images against source prompts. Applied rubric-based assessments to measure semantic alignment, artifact detection, compliance, and overall output quality. Used these rankings to improve visual model behavior and output reliability. • Ran A/B testing across computer vision pipelines. • Performed rubric-based rankings of generated images vs source prompts. • Evaluated semantic alignment, artifacts, compliance, and output quality. • Supported QA-driven iteration of image generation performance.

2025 - Present

Software Developer/AI Trainer — Speech ASR-TTS validation for SFT and optimization

AudioTranscription

Validated speech ASR-TTS model outputs by comparing audio analysis against expected transcription and intent signals for supervised fine-tuning readiness. Assessed transcription accuracy, waveform clarity, and intent preservation to identify issues and enable performance optimizations. Used evaluation results to support iterative model improvements. • Compared audio analysis for transcription and intent preservation. • Evaluated transcription accuracy and waveform clarity. • Fed validation insights into SFT support and performance optimization. • Performed quality checks aligned with model benchmarking goals.

2025 - Present

Software Developer/AI Trainer — LLM validation, red teaming, HITL evaluation, and RLHF-aligned benchmarking

TextRLHFRed Teaming

Performed LLM output validation and safety alignment auditing to support scalable AI reliability and evaluation workflows. Conducted Human-in-the-Loop (HITL) evaluations and RLHF-aligned model benchmarking using prompt-engineered conversational simulations to assess instruction adherence, grounding, latency, and coherence. Also supported supervised fine-tuning and reliability improvements by reviewing model responses for factual accuracy, hallucinations, bias mitigation, and policy compliance. • Audited responses for factual accuracy, hallucination detection, bias mitigation, and policy compliance. • Ran HITL and RLHF-aligned benchmarking via structured conversational simulations. • Evaluated instruction adherence, contextual grounding, response latency, and dialog coherence. • Produced high-quality feedback to improve model performance and user expectations.

2025 - Present

Education

Work History