Prompt Engineer
Zensar Technologies · Oct 2024 – Present
- Design, test, and iterate on prompts to improve LLM output quality, running structured evaluation suites that surface edge-case failures and drive measurable gains in accuracy and consistency across locales.
- Evaluate and optimize AI-generated code and model outputs for accuracy, efficiency, and enterprise-grade reliability across complex programming and reasoning tasks as a human-in-the-loop LLM evaluator.
- Perform quality assurance and human evaluation of AI-driven solutions against enterprise software standards, partnering cross-functionally to turn evaluation findings into prompt and rule improvements.
- Build and maintain text normalization (TN) and inverse text normalization (ITN) grammars for enterprise speech AI (ASR/TTS) pipelines, covering dates, times, currencies, measurements, and domain-specific formats.

