Brendan Keller

FAQ About Brendan Keller

• What is Brendan Keller's address?

  Brendan Keller is located at
  1551 4th AvenueUnit 802, San Diego, United States

• What is Brendan Keller's job title?

  Brendan Keller is a
  Senior data scientist with eight years of machine learning and risk management experience. PhD in computational physics. Familiar with end-to-end development and deployment of both online and offline Python machine learning and AI models in the IT, finance and insurance space. Also has hands on experience with identifying complex data features that have been used by business units to identify and avoid bad risk. Interfaced with multiple stakeholders in order to push machine learning & data science projects into day to day operations so they can mitigate risk or save costs. Examples of value I've brought to past machine learning/AI projects: -Developed and deployed a machine learning model that classified insurance policy risk at Hippo. This involved everything from optimizing the model, selecting features, cleaning and processing the data, working with business to set thresholds, working with MLops to integrate the model into the existing engineering pipeline and QA to ensure proper behavior when deployed. This new model had a smooth deployment and was being used by underwriting as soon as it was deployed. -A customer gave us multiple datasets for a project, including a very rich geolocation dataset. FICO traditionally only uses the transaction data to build neural networks and this geolocation data did not fit this traditional mold so it was originally excluded. I came up with an approach using latent dirichelet allocation to integrate the information in this dataset into a new fraud detection variable and train a neural network on it in order to improve our fraud detection ability. ✦Programming: Python, Java, C++, Bash, Perl, R, multithreading. ✦ Machine Learning: Neural Networks (Deep Learning), Regression, Clustering, Scikit-Learn, PyTorch, Decision Trees, Random Forests, XGBoost, PySpark, Transformers, Large Language Models. ✦ Big Data: SQL, MySQL, BigQuery, Apache Spark, Hadoop, Pandas. ✦ Statistical Modeling: Bayesian Statistics, Hypothesis Testing, Pearson’s Correlations, Digital Filters, Markov Chains, A/B testing. ✦ Data Visualization: Matplotlib, Bokeh, Superset, Tableau, Seaborn. ✦ Cloud CompuBng: Google Cloud, AWS, Azure. ✦ Version Control and Web App: GitHub, Subversion, Flask, Bokeh, HTML, XML. Fico

• What industry does Brendan Keller currently work in?

  Brendan Keller currently works in the
  Fico

• What is Brendan Keller's area of expertise?

  Brendan Keller specializes in
  Senior data scientist with eight years of machine learning and risk management experience. PhD in computational physics. Familiar with end-to-end development and deployment of both online and offline Python machine learning and AI models in the IT, finance and insurance space. Also has hands on experience with identifying complex data features that have been used by business units to identify and avoid bad risk. Interfaced with multiple stakeholders in order to push machine learning & data science projects into day to day operations so they can mitigate risk or save costs. Examples of value I've brought to past machine learning/AI projects: -Developed and deployed a machine learning model that classified insurance policy risk at Hippo. This involved everything from optimizing the model, selecting features, cleaning and processing the data, working with business to set thresholds, working with MLops to integrate the model into the existing engineering pipeline and QA to ensure proper behavior when deployed. This new model had a smooth deployment and was being used by underwriting as soon as it was deployed. -A customer gave us multiple datasets for a project, including a very rich geolocation dataset. FICO traditionally only uses the transaction data to build neural networks and this geolocation data did not fit this traditional mold so it was originally excluded. I came up with an approach using latent dirichelet allocation to integrate the information in this dataset into a new fraud detection variable and train a neural network on it in order to improve our fraud detection ability. ✦Programming: Python, Java, C++, Bash, Perl, R, multithreading. ✦ Machine Learning: Neural Networks (Deep Learning), Regression, Clustering, Scikit-Learn, PyTorch, Decision Trees, Random Forests, XGBoost, PySpark, Transformers, Large Language Models. ✦ Big Data: SQL, MySQL, BigQuery, Apache Spark, Hadoop, Pandas. ✦ Statistical Modeling: Bayesian Statistics, Hypothesis Testing, Pearson’s Correlations, Digital Filters, Markov Chains, A/B testing. ✦ Data Visualization: Matplotlib, Bokeh, Superset, Tableau, Seaborn. ✦ Cloud CompuBng: Google Cloud, AWS, Azure. ✦ Version Control and Web App: GitHub, Subversion, Flask, Bokeh, HTML, XML.

• Which languages does Brendan Keller speak? 

  Brendan Keller speaks the following languages:
  No languages available.

• Which industries has Brendan Keller previously worked in?   

  Brendan Keller has prior experience in the following industries:
  Fico , Jolata , Uc Santa Barbara , Ucsb Department Of Physics , The Data Incubator

• What is Brendan Keller’s work experience?  

  Brendan Keller has the following professional experience:
  Fico , Fico , Jolata , Uc Santa Barbara , Ucsb Department Of Physics , The Data Incubator .

• How can I find Brendan Keller on LinkedIn?  

  You can view Brendan Keller’s
  LinkedIn profile at LinkedIn URL.

• What are Brendan Keller’s work email addresses?  

  Brendan Keller’s work email addresses are
  [email protected] .

• What company does Brendan Keller work for?  

  Brendan Keller works at
  Fico