George Gardere

Key Skills

Software

Data Annotation Tech

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code Programming

Top Task Types

Computer Programming/Coding

Fine-tuning

Freelancer Overview

I have done extensive ai training and data annotation work spanning across multiple project lines, mainly focusing on improving and validating model generated codebases. From significant experience writing and refining code, evaluating model outputs compared to baselines, and modifying where needed to ensure the full functionality and as close to "perfect" results as possible. Skilled at developing complex prompts to stress test models, doing side by side comparisons against baseline outputs, and applying high and low level evaluation criteria to both codebases to assess accuracy, and ensure functionality, and quality. I've also built QA workflows to validate model outputs in real environments by running models through Dockerized test setups, and integrating ai generated outputs into existing, active apps and UIs. I refined system files in the environment to ensure smooth execution.

Labeling Experience

Prompt and Criteria-based model evaluation

Data Annotation TechTextEvaluation RatingRed Teaming

Designed high and low level evaluation criteria to review Ai model responses. Created complex prompts to stress test models. Measured and evaluated model performance in different scenarios, documenting results to highlight areas in need of refinement. Contributed to building feedback pipelines to improve model reasoning and quality of output.

Designed high and low level evaluation criteria to review Ai model responses. Created complex prompts to stress test models. Measured and evaluated model performance in different scenarios, documenting results to highlight areas in need of refinement. Contributed to building feedback pipelines to improve model reasoning and quality of output.

2025

Ai code validation in Docker environments

Data Annotation TechComputer Code ProgrammingEvaluation RatingComputer Programming Coding

Executed Ai code evaluation pipelines in Dockerized environments, validating outputs by running all codebases (original/preedit, target model, baseline model, and refined rewrite) through a given test suite with my own unit tests. Modified the Docker build and run configs to integrate the codebases smoothly. Overall, focused on ensuring ease in reproducing and giving thorough QA across sys setups.

Executed Ai code evaluation pipelines in Dockerized environments, validating outputs by running all codebases (original/preedit, target model, baseline model, and refined rewrite) through a given test suite with my own unit tests. Modified the Docker build and run configs to integrate the codebases smoothly. Overall, focused on ensuring ease in reproducing and giving thorough QA across sys setups.

2025

Ai code evaluation and refinement

Data Annotation TechComputer Code ProgrammingText GenerationEvaluation Rating

Reviewed ai generated code and compared model output to baselines to evaluate accuracy, and functionality. This includes side by side comparisons, refining model outputs to ensure final copy is fully functional, and making sure it is all according to provided rubrics/requirements. Created prompts that test edge cases, and identified weaknesses in model logic and processes.

Reviewed ai generated code and compared model output to baselines to evaluate accuracy, and functionality. This includes side by side comparisons, refining model outputs to ensure final copy is fully functional, and making sure it is all according to provided rubrics/requirements. Created prompts that test edge cases, and identified weaknesses in model logic and processes.

2024

Education

Work History