Daniel Higley, Phd work email
- Valid
- Valid
- Valid
Daniel Higley, Phd personal email
Daniel Higley, Phd phone numbers
Developing machine learning models and machine learning-powered applications
-
Senior Data ScientistFinraNew York, Ny, Us
-
Senior Data ScientistFinra Sep 2021 - PresentDeveloped machine learning models and the applications that use them- Developed an anomaly-based network intrusion detection model and surrounding application for use by cybersecurity department. The application included SHAP explanations of detected anomalies for users to quickly understand why instances were considered anomalous by the model and act accordingly.- Led transformation of model training processes from monolithic, implicitly-connected Jupyter notebooks with significant manual processes into end-to-end automated, auditable and reproducible model training pipelines.- Built a Python backend that used Retrieval Augmented Generation (RAG) with a Large Language Model (LLM) for question answering based on financial documents (1st place Createathon project in LLM category).- Improved precision of an algorithm by 15 percent (without a drop in recall) by changing data processing to better identify entities of interest.Founded and developed model validation practice for machine learning-based algorithmic patterns used to detect prohibited behaviors in financial markets (model validation practice later integrated with development as single team)- Developed framework to ensure present and future ML model performance and maintainability- Onboarded and assisted several colleagues new to model validation- Performed model validation for many machine learning models (>10)- Discovered opportunities for improvement in model reproducibility, feature preparation, linkage of training data to business use case, model maintainability and complexity reduction-Tuned model performance measures to better align with business objectives -
Data ScientistFinra Feb 2021 - Sep 2021
-
Research AssociateSlac National Accelerator Laboratory May 2018 - Jul 2020Menlo Park, California, United States- Developed a new algorithm for signal deconvolution for use as part of a method to simplify and increase the sensitivity of a common X-ray spectroscopy. Created a Python package for others to use this method, which includes the developed algorithm as a class that conforms to the Scikit-Learn Estimator API (see github.com/dhigley6/PAX2).- Collected and analyzed experimental data -
X-Ray Spectroscopy Researcher And Data AnalystStanford University Sep 2012 - Apr 2018Stanford, CaDevelopment and use of new X-ray spectroscopy techniques which leveraged the extremely high intensity and short pulse duration LCLS X-ray laser facility. I spent the majority of my time analyzing and visualizing the many Terabytes of data which our experiments produced.- Led three large-scale experiments and analyzed the Terabytes of image, time series and structured data each experiment produced. One experiment demonstrated how to increase the efficiency by more than a million times of a common X-ray characterization technique.- Developed a real-time data visualization dashboard and instrumentation enabling high resolution and accuracy femtosecond time- and polarization-resolved X-ray absorption spectroscopy. The dashboard calculated and presented diagnostic information on streaming data (https://github.com/dhigley6/LCLS_XAS_online). This technology has since been used by several different teams, resulting in publications in top journals.-Rapidly analyzed data for more than 18 several-day-long experiments at large-scale scientific facilities (analysis deadlines ranged from an hour to a day after data collection). Experiments focussed on understanding ultrafast dynamics in materials. Better understanding of these dynamics could be used to develop better computers.- Performed custom numerical simulations of femtosecond material dynamics- Designed, assembled, and tested ultrahigh vacuum X-ray and electron experimental apparatus -
Optics And Microwave Engineering Researcher And Data AnalystColorado State University Oct 2009 - Sep 2012Fort Collins, Colorado Area- Computationally and experimentally analyzed the effects of diffraction and aberration on a computational imaging technique (SPIFI, Spatially Integrated Chirped Frequency Modulation Imaging)- Analyzed microwave radiometer and optical biosensor data- Developed a research group website -
TutorColorado State University Sep 2009 - Jan 2010Fort Collins, Colorado Area- Tutored students in math, physics and electrical engineering courses
-
Microwave Engineering Research AssistantJet Propulsion Laboratory Jun 2010 - Aug 2010Greater Los Angeles Area- Analyzed and recorded high-frequency microwave radiometer testbed data
Daniel Higley, Phd Skills
Daniel Higley, Phd Education Details
-
Applied Physics
-
Electrical Engineering
Frequently Asked Questions about Daniel Higley, Phd
What company does Daniel Higley, Phd work for?
Daniel Higley, Phd works for Finra
What is Daniel Higley, Phd's role at the current company?
Daniel Higley, Phd's current role is Senior Data Scientist.
What is Daniel Higley, Phd's email address?
Daniel Higley, Phd's email address is dh****@****ord.edu
What is Daniel Higley, Phd's direct phone number?
Daniel Higley, Phd's direct phone number is +130325*****
What schools did Daniel Higley, Phd attend?
Daniel Higley, Phd attended Stanford University, Colorado State University.
What are some of Daniel Higley, Phd's interests?
Daniel Higley, Phd has interest in Condensed Matter Physics, Microscopy, Optical Physics.
What skills is Daniel Higley, Phd known for?
Daniel Higley, Phd has skills like Optics, Matlab, Experimentation, R, Microscopy, Photonics, Condensed Matter Physics, Research, Python, Materials Science, Data Analysis, Nanotechnology.
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial