Anusha Janagam

Prompt Generalist @Soul Ai

Currently evaluating large language model (LLM) responses to diverse prompts, ensuring high-quality and contextually relevant outputs.• Providing continuous feedback to improve LLM accuracy and specialising in prompt engineering, NLP evaluation frameworks.• Contributing to Project Shield – a side-by-side evaluation initiative comparing responses from two AI agents to benchmark performance on reasoning, relevance, fluency, and safety.• Assisting in identifying hallucinations, biases, and inconsistencies to enhance the overall reliability and usability of AI outputs.