Portfolio ↗ · LinkedIn ↗ · Say hello ↗
I'm Justin, a Machine Learning Engineer in Search at TikTok. My work connects retrieval and ranking model design with evaluation, online serving, and experimentation.
- Retrieval & ranking: Generative retrieval from dataset construction through training, indexing, and serving; integration with multi-stage ranking systems.
- Model scaling & learning objectives: Transformer rankers, warm-start strategies for sparse embeddings and dense components, listwise distillation, and calibrated multi-task learning.
- Production reliability: Offline-to-online feature consistency, ranking observability, model-launch validation, and online experiments.
Previously, I worked on Bayesian time-series modeling and multimodal anomaly detection at Argonne National Laboratory. I'm a CMU MISM graduate and UC San Diego alumnus.
Away from models: basketball, fitness, cooking, and exploring trails. Still curious, innovative, and motivated :)
Let's connect. I'm open to collaborating on LLM and Generative AI projects. Find me on LinkedIn, explore my portfolio, or reach out at zechengl@alumni.cmu.edu.



