Hire Scott B. - AI Researcher

Lead R&D for government projects, focusing on coding and implementing object detection models capable of real-time inference on high-resolution (5k x 5k) images. Develop robust and scalable distributed computer vision systems by experimenting with and evaluating models for various tasks, including: - Segmentation - Saliency detection - Object detection and tracking - Representation learning - Fine-grained classification Specialize in the implementation of advanced ML tracking algorithms for visual and infrared imaging, utilizing techniques such as Kalman filters, UKFs, particle filters, Gaussian-sum filters, and track-before-detect. Experiment with smart data association methods to enhance multi-object tracking in uncertain and low-contrast scenarios. Conduct extensive model training and evaluation to optimize performance, iterating on algorithms and code based on experimental results. Collaborate with implementation engineers to fine-tune model code for edge hardware, ensuring blazingly fast execution. Set up and manage MLOps infrastructure to monitor training and inference results, integrating the latest advancements from top research and conferences like CVPR and NeurIPS. Regularly perform literature reviews to stay current with state-of-the-art techniques, applying new findings to improve model implementation and performance. Serve as the primary AI/ML expert across the organization, participating in interviews for new team members. Engage in all aspects of model development, from coding and experimenting with novel algorithms to comprehensive training and evaluation.

Design, develop, and deliver solutions based on the client's needs. Lead team of Data Visualization and Machine Learning engineers in developing custom tooling for executives to make more informed business decisions Utilize unsupervised ensemble machine learning with Tensorflow and AWS Sagemaker to guide corporate decisions and provide useful correlations and information. Incorporate new techniques in Bayesian modelling (probabilistic neural networks) to provide measurable uncertainty in prediction and increase transparency - thanks arXiv! Investigated new and emerging software applications within Machine Learning to stay on the cutting edge in terms of best practices and techniques Develop CRM email sync application using AWS Lambda, DynamoDB, Kubernetes, and Microsoft Graph to synchronize millions of contacts to Office 365 Develop custom integrations with multiple 3rd party software vendors to automatically create accounts for new employees Help guide junior developers in learning Go, Docker, and Kubernetes.