Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Staff Data Scientist with 8 years of experience designing and deploying production ML systems. I build scalable data pipelines and cloud-native inference services using Python, PyTorch, TensorFlow, Docker, and Kubernetes—optimizing latency, reliability, and performance. I also architect enterprise GenAI and agentic AI platforms with secure RAG and LLMOps governance. My work emphasizes model monitoring, versioning, reproducibility, and auditability across AWS and GCP to drive measurable product outcomes.…Staff Data Scientist with 8 years of experience designing and deploying production ML systems. I build scalable data pipelines and cloud-native inference services using Python, PyTorch, TensorFlow, Docker, and Kubernetes—optimizing latency, reliability, and performance. I also architect enterprise GenAI and agentic AI platforms with secure RAG and LLMOps governance. My work emphasizes model monitoring, versioning, reproducibility, and auditability across AWS and GCP to drive measurable product outcomes.

Kenan Herbert

Data Scientist, AI Engineer, Data Engineer, +2





Staff Data Scientist with 8 years of experience designing and deploying production ML systems. I build scalable data pipelines and cloud-native inference services using Python, PyTorch, TensorFlow, Docker, and Kubernetes—optimizing latency, reliability, and performance. I also architect enterprise GenAI and agentic AI platforms with secure RAG and LLMOps governance. My work emphasizes model monitoring, versioning, reproducibility, and auditability across AWS and GCP to drive measurable product outcomes.…Staff Data Scientist with 8 years of experience designing and deploying production ML systems. I build scalable data pipelines and cloud-native inference services using Python, PyTorch, TensorFlow, Docker, and Kubernetes—optimizing latency, reliability, and performance. I also architect enterprise GenAI and agentic AI platforms with secure RAG and LLMOps governance. My work emphasizes model monitoring, versioning, reproducibility, and auditability across AWS and GCP to drive measurable product outcomes.

Available to hire

Staff Data Scientist with 8 years of experience designing and deploying production ML systems. I build scalable data pipelines and cloud-native inference services using Python, PyTorch, TensorFlow, Docker, and Kubernetes—optimizing latency, reliability, and performance.

I also architect enterprise GenAI and agentic AI platforms with secure RAG and LLMOps governance. My work emphasizes model monitoring, versioning, reproducibility, and auditability across AWS and GCP to drive measurable product outcomes.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Beginner

Work Experience

Staff Data Scientist at Checkr, Inc

May 1, 2025 - Present

Implemented production-grade PyTorch transformer architectures and Kubernetes microservices for distributed inference, improving scalability, reliability, observability, and enterprise model governance. Built security-enhanced enterprise GenAI systems with agentic AI and secure RAG, including governance and auditability across production knowledge repositories and infrastructure. Established LLMOps lifecycle administration: versioning, evaluation infrastructure, performance drift detection, and automated rollout strategies across Kubernetes with AWS/GCP. Developed production Python and SQL pipelines processing terabytes for feature generation and real-time API infrastructure. Integrated Dockerized TensorFlow/PyTorch models into cloud-native backend services for low-latency predictions. Coordinated with product/data/engineering teams on design and code reviews, and implemented end-to-end MLOps governance for reproducibility, observability, auditability, and compliance.

Senior Data Scientist at Truework

May 1, 2019 - May 1, 2025

Built production ML models in PyTorch and TensorFlow for employment and income verification with high accuracy at worldwide scale. Implemented model monitoring and versioning using custom pipelines and Kubernetes to ensure reproducible training, lineage transparency, and automated alerts for continuous performance degradation detection. Designed SQL/Python data pipelines ingesting millions of records to support feature engineering and automated model retraining for high-volume real-time verification APIs. Deployed Dockerized microservices on AWS with REST endpoints to reduce prediction latency and improve scalability. Optimized deep learning inference with quantization and batching to cut infrastructure costs while maintaining performance. Collaborated with engineering/product teams to drive adoption of MLOps practices and production readiness across multiple services, and mentored junior data scientists through code reviews and deployment best practices.

Associate Data Scientist at Zendrive

May 1, 2018 - April 1, 2019

Engineered Python pipelines for large-scale telematics sensor data, cleaning and transforming millions of driving events to enable feature extraction for model training. Built predictive PyTorch models for driver risk scoring, iterating on feature engineering and evaluating with precision/recall/lift to drive business improvements. Automated continuous model evaluation workflows using Python to reduce manual effort and speed experimentation. Partnered on code reviews and documentation to maintain reproducibility and engineering standards. Deployed results into interactive analytics dashboards using SQL and Python for stakeholder visibility into model behavior and data quality. Supported migration of training workflows from on-prem to AWS. Conducted ad-hoc statistical analyses to identify accident-probability patterns and assisted with data validation checks and automated tests for pipeline reliability.