ML Engineer
- Benchmarked and evaluated LLMs and multimodal generative models (video, image, audio) across proprietary and open-source systems, including Google Veo
- Designed and tested pipelines for long-form AI video generation, focusing on story, character, and style consistency
- Experimented with RAG and fine-tuning techniques to improve narrative coherence and output stability
- Documented results using structured benchmarks and identified feasible approaches for AI-powered content creation workflows

