LLM Fine-tuning and Evaluation Lead (Software Engineer)
As a Software Engineer at After Query, I led the training and fine-tuning of specialized LLaMA-based Large Language Models for benchmarking and insight generation. I authored solution scripts and backend logic to provide ground-truth paths, and created instruction sets with a realistic writing style. I validated agent outputs and transformed raw telemetry into structured data for agent testing. • Designed test suites for strict agent schema validation. • Authored natural language instructions for LLM training scenarios. • Led finetuning of LLMs for predictive analytics. • Transformed telemetry into structured data for model evaluation.