For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
T
Tamara Laws

Tamara Laws

AI Training-Prompt Writer

USA flagGreenville, Usa
$14.00/hrIntermediateOtherMercor

Key Skills

Software

Other
MercorMercor

Top Subject Matter

General Domain Expertise
Business Domain Expertise
Management Domain Expertise

Top Data Types

AudioAudio
DocumentDocument
ImageImage
TextText
VideoVideo

Top Task Types

Audio RecordingAudio Recording
Bounding BoxBounding Box
Object DetectionObject Detection
Prompt + Response Writing (SFT)Prompt + Response Writing (SFT)
Question AnsweringQuestion Answering
RelationshipRelationship
TranscriptionTranscription
Translation/LocalizationTranslation/Localization

Freelancer Overview

I have extensive experience in the full lifecycle of AI training data, specializing in both multimodal data annotation and complex prompt engineering. My background includes high-precision work in Computer Vision, where I performed intricate tasks such as object removal, temporal video editing, and semantic tagging of products and entities across various media. This hands-on experience with diverse data types—including text, audio, and video—has given me a deep understanding of how to curate the high-fidelity ground truth data necessary for training robust, versatile models. Beyond standard labeling, I specialize in architecting end-to-end, multi-turn conversational datasets designed to enhance model reasoning and instruction-following capabilities. I am adept at identifying edge cases and developing sophisticated prompt-response scenarios from scratch to support Supervised Fine-Tuning (SFT) and RLHF workflows. My ability to maintain logical consistency across complex interactions, combined with a rigorous approach to data quality, allows me to produce high-impact training materials that significantly improve model performance and alignment.

IntermediateEnglish

Labeling Experience

Prompt Writer

DocumentPrompt Response Writing SFT
Designed and developed complex, synthetic datasets to facilitate the Supervised Fine-Tuning (SFT) of Large Language Models. The project scope focused on creating sophisticated, multi-turn conversational architectures that challenged model reasoning, instruction-following, and contextual memory. Specific Data Labeling Tasks Performed: Authored end-to-end multi-turn dialogues from scratch, simulating complex user-AI interactions. Developed "edge-case" prompts requiring the model to navigate conflicting instructions or maintain a specific persona over long-form conversations. Generated high-quality ground-truth responses to train models in specific reasoning tasks and logical deduction. Quality Measures Adhered To: Followed rigorous rubrics for linguistic precision, factual accuracy, and logical consistency. Adhered to strict alignment guidelines regarding model safety, tone, and helpfulness. Participated in iterative feedback cycles to ensure data met high-complexity benchmarks for model training. 2. Outlier (Generalist Data Annotation) Project Description: Contributed to the optimization of multimodal AI models through high-fidelity data refinement and Computer Vision (CV) tasks. The project scope involved improving model perception and object-handling capabilities across various media formats, including video, image, and audio. Specific Data Labeling Tasks Performed: Executed complex video manipulation tasks, including removing specific items and using text prompts to direct the model in re-inserting objects (inpainting). Performed entity tagging and semantic labeling of products, people, and landmarks to enhance object detection datasets. Transcribed audio and video content to provide ground-truth data for speech-to-text and natural language processing models. Quality Measures Adhered To: Maintained 95%+ accuracy ratings by strictly adhering to granular, project-specific style guides. Ensured temporal consistency in video tasks to minimize artifacts and training noise. Collaborated within a high-volume environment while meeting strict deadlines and inter-annotator agreement standards.

Designed and developed complex, synthetic datasets to facilitate the Supervised Fine-Tuning (SFT) of Large Language Models. The project scope focused on creating sophisticated, multi-turn conversational architectures that challenged model reasoning, instruction-following, and contextual memory. Specific Data Labeling Tasks Performed: Authored end-to-end multi-turn dialogues from scratch, simulating complex user-AI interactions. Developed "edge-case" prompts requiring the model to navigate conflicting instructions or maintain a specific persona over long-form conversations. Generated high-quality ground-truth responses to train models in specific reasoning tasks and logical deduction. Quality Measures Adhered To: Followed rigorous rubrics for linguistic precision, factual accuracy, and logical consistency. Adhered to strict alignment guidelines regarding model safety, tone, and helpfulness. Participated in iterative feedback cycles to ensure data met high-complexity benchmarks for model training. 2. Outlier (Generalist Data Annotation) Project Description: Contributed to the optimization of multimodal AI models through high-fidelity data refinement and Computer Vision (CV) tasks. The project scope involved improving model perception and object-handling capabilities across various media formats, including video, image, and audio. Specific Data Labeling Tasks Performed: Executed complex video manipulation tasks, including removing specific items and using text prompts to direct the model in re-inserting objects (inpainting). Performed entity tagging and semantic labeling of products, people, and landmarks to enhance object detection datasets. Transcribed audio and video content to provide ground-truth data for speech-to-text and natural language processing models. Quality Measures Adhered To: Maintained 95%+ accuracy ratings by strictly adhering to granular, project-specific style guides. Ensured temporal consistency in video tasks to minimize artifacts and training noise. Collaborated within a high-volume environment while meeting strict deadlines and inter-annotator agreement standards.

2026 - 2026

Generalist Data Annotator

AudioTranscription
. Outlier (Transcription) Labeling Type: Transcription Data Type: Audio Project Description: Provided high-accuracy ground-truth text for auditory media to improve speech recognition and multimodal model comprehension. Scope of Project: Transcribed audio and video dialogue into structured text formats, ensuring all linguistic nuances were captured correctly. Specific Data Labeling Tasks Performed: Performed verbatim transcription of dialogue and environmental sounds, including speaker identification and timestamping for audio-visual alignment. Quality Measures Adhered To: Adhered to strict orthographic standards and achieved a 98%+ accuracy rate on complex audio files with varied accents and background noise.

. Outlier (Transcription) Labeling Type: Transcription Data Type: Audio Project Description: Provided high-accuracy ground-truth text for auditory media to improve speech recognition and multimodal model comprehension. Scope of Project: Transcribed audio and video dialogue into structured text formats, ensuring all linguistic nuances were captured correctly. Specific Data Labeling Tasks Performed: Performed verbatim transcription of dialogue and environmental sounds, including speaker identification and timestamping for audio-visual alignment. Quality Measures Adhered To: Adhered to strict orthographic standards and achieved a 98%+ accuracy rate on complex audio files with varied accents and background noise.

2025 - 2026

Generalist Data Annotator

ImageEntity Ner Classification
Outlier (Tagging/Classification) Labeling Type: Classification / Entity NR classification Data Type: Image Project Description: Identified and categorized key visual entities within large-scale datasets to enhance model recognition of real-world objects and locations. Scope of Project: Tagged a high volume of media to help the model accurately differentiate between products, people, and specific places. Specific Data Labeling Tasks Performed: Applied classification labels and Named Entity Recognition (NR) to products, landmarks, and individuals in diverse photos and video stills. Quality Measures Adhered To: Followed granular taxonomy guidelines and maintained high inter-annotator agreement to minimize noise in the training data.

Outlier (Tagging/Classification) Labeling Type: Classification / Entity NR classification Data Type: Image Project Description: Identified and categorized key visual entities within large-scale datasets to enhance model recognition of real-world objects and locations. Scope of Project: Tagged a high volume of media to help the model accurately differentiate between products, people, and specific places. Specific Data Labeling Tasks Performed: Applied classification labels and Named Entity Recognition (NR) to products, landmarks, and individuals in diverse photos and video stills. Quality Measures Adhered To: Followed granular taxonomy guidelines and maintained high inter-annotator agreement to minimize noise in the training data.

2025 - 2026

Generalist Data Annotator

VideoRLHF
Outlier (Video Editing/Inpainting) Labeling Type: Object detection / RLHF Data Type: Video Project Description: Performed advanced video manipulation and evaluated generative model responses to improve object removal and text-to-video editing. Scope of Project: Identified and removed specific items (like furniture) from video footage and utilized prompts to guide the AI in replacing those items realistically. Specific Data Labeling Tasks Performed: Executed object identification for removal and provided Reinforcement Learning from Human Feedback (RLHF) by rating the model’s accuracy in re-inserting objects based on text prompts. Quality Measures Adhered To: Maintained strict temporal consistency and visual logic; ensured no artifacts were present in the final generative output.

Outlier (Video Editing/Inpainting) Labeling Type: Object detection / RLHF Data Type: Video Project Description: Performed advanced video manipulation and evaluated generative model responses to improve object removal and text-to-video editing. Scope of Project: Identified and removed specific items (like furniture) from video footage and utilized prompts to guide the AI in replacing those items realistically. Specific Data Labeling Tasks Performed: Executed object identification for removal and provided Reinforcement Learning from Human Feedback (RLHF) by rating the model’s accuracy in re-inserting objects based on text prompts. Quality Measures Adhered To: Maintained strict temporal consistency and visual logic; ensured no artifacts were present in the final generative output.

2025 - 2026

Education

G

Greenville Technical College

Associate's Degree, Business Management

Associate's Degree
2022 - 2025
G

Greenville Technical College

No Degree Awarded, Medical Laboratory Technology

No Degree Awarded
2013 - 2015

Work History

P

Precision Horizontal Boring

Office Manager

Taylors
2022 - 2025
S

Spectrum

Customer Service Representative

Simpsonville
2019 - 2023