Veera Chundru

Key Skills

Software

AWS SageMaker

Labelbox

SuperAnnotate

Other

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Text

Top Label Types

Prompt Response Writing SFT

RLHF

Evaluation Rating

Classification

Fine Tuning

Freelancer Overview

I have hands-on experience evaluating the scientific correctness, conceptual rigor, and depth of large language model outputs across advanced physics and engineering domains. My work involved identifying factual inaccuracies, reasoning gaps, and conceptual misunderstandings in model-generated responses, as well as benchmarking model performance on complex physics problems and applied engineering scenarios. I worked independently using proprietary evaluation and annotation tools, ensuring high-quality, reliable feedback to improve model reasoning and accuracy. In addition to model evaluation, I have extensive experience extracting and structuring information from engineering drawings, technical reports, and detailed design specifications. I annotated datasets capturing key design variations and optimization parameters, enabling data-driven improvements in system performance, including projects focused on enhancing moisture separation efficiency. This work required precision, domain understanding, and consistency. Academically, I hold a Master’s degree in Robotics and a Bachelor’s degree in Mechanical Engineering, providing a strong foundation in applied mathematics, physics, and engineering systems. I also worked for four years as a Research Assistant on robot path planning for cold spray manufacturing, where I combined algorithmic reasoning with real-world manufacturing constraints. Additionally, I completed a mechanical design optimization project focused on moisture separator performance, reinforcing my ability to translate complex engineering concepts into structured, evaluable data. Together, these experiences position me well for AI trainer roles requiring deep technical expertise, analytical rigor, and high-quality human-in-the-loop evaluation.

Labeling Experience

Mechanical Engineering expert

SuperannotateTextRLHFEvaluation Rating

Extracted and structured information from engineering drawings, technical reports, and design specifications. Annotated datasets capturing design variations and optimisation parameters to enable data-driven improvement of moisture separation efficiency. Evaluated scientific correctness, conceptual rigour, and depth of LLM-generated responses across physics and engineering topics.

Extracted and structured information from engineering drawings, technical reports, and design specifications. Annotated datasets capturing design variations and optimisation parameters to enable data-driven improvement of moisture separation efficiency. Evaluated scientific correctness, conceptual rigour, and depth of LLM-generated responses across physics and engineering topics.

2025

Physics Labeling Expert

Aws SagemakerTextPrompt Response Writing SFT

Evaluated scientific correctness, conceptual rigour, and depth of LLM-generated responses across physics and engineering topics. Identified factual errors, reasoning flaws, and conceptual misunderstandings in model outputs. Benchmarked model performance on advanced physics and applied engineering problems. Worked independently using proprietary evaluation tools.

Evaluated scientific correctness, conceptual rigour, and depth of LLM-generated responses across physics and engineering topics. Identified factual errors, reasoning flaws, and conceptual misunderstandings in model outputs. Benchmarked model performance on advanced physics and applied engineering problems. Worked independently using proprietary evaluation tools.

2025

STEM Expert

OtherTextClassificationRLHF

Classified and labeled physics and mechanical engineering data for LLM training, ensuring technical and mathematical accuracy. Created and validated expert training examples to support fine-tuning on applied physics and engineering reasoning. Wrote and refined domain-specific prompts to evaluate and stress-test model reasoning on complex problems. Assessed LLM outputs for correctness and coherence, providing structured feedback to reduce technical errors.

Classified and labeled physics and mechanical engineering data for LLM training, ensuring technical and mathematical accuracy. Created and validated expert training examples to support fine-tuning on applied physics and engineering reasoning. Wrote and refined domain-specific prompts to evaluate and stress-test model reasoning on complex problems. Assessed LLM outputs for correctness and coherence, providing structured feedback to reduce technical errors.

2025 - 2025

Education

Work History