Sai Bhavani

Operations Associate

Hyderabad, Telangana, India1 yr 1 mo experience

Key Highlights

Expert in validating autonomous vehicle CoT outputs.
Strong collaboration with prompt engineers to enhance model accuracy.
Proficient in keyframe identification and event labeling.

Stackforce AI infers this person is a Quality Analyst specializing in Autonomous Vehicle technology.

Contact

Skills

Core Skills

Quality AnalysisKeyframe Identification & Event LabelingSql

Other Skills

Oracle DatabaseScenario InterpretationLLM PromptingCoT Reasoning ValidationHTML5Object-Oriented Programming (OOP)Data ManagementAnalytical SkillsProblem SolvingData AnalyticsC (Programming Language)Python (Programming Language)PL/SQL

About

Quality Analyst with hands-on experience in Autonomous Vehicle Chain-of-Thought (CoT) evaluation, AI reasoning validation, and prompt-based scenario analysis. I specialize in reviewing driving clips, verifying keyframes, event labels, and ensuring that model decisions align with safety and ODD guidelines. I work closely with prompt engineers and SMEs to refine CoT logic, improve model accuracy, and maintain high-quality annotations for AV systems. I support quality tracking, error identification, and continuous improvement initiatives.

Experience

1 yr 1 mo

Total Experience

1 yr 1 mo

Average Tenure

1 yr 1 mo

Current Experience

Nvidia

2 roles

Process Executive

Promoted

May 2025 – Present · 1 yr 1 mo

Performed quality validation of Chain-of-Thought (CoT) outputs for autonomous driving scenarios to ensure accuracy, consistency, and compliance with safety guidelines.
Reviewed and verified keyframes, event timing, classifications, and vehicle interactions in driving clips.
Identified errors in model-generated reasoning and ensured alignment with ODD (Operational Design Domain) specifications.
Collaborated with prompt engineers and SMEs to refine guidelines, reduce error rates, and improve model safety and accuracy.

Keyframe Identification & Event LabelingQuality Analysis

Associate Process Executive

Feb 2025 – May 2025 · 3 mos

Contributed to an autonomous driving project focused on annotating wait
conditions, traffic signs, and traffic light states across diverse driving
scenarios.
Identified and validated traffic signal states (red, yellow, green), sign compliance,
and ego-vehicle behavioral responses in both simple and complex intersections.
Ensured high-quality annotations by reviewing scene transitions, temporal
consistency, and driver intent patterns across frames.