About Me
Dillon Camp is a Data DevOps Engineer at Anvilogic, specializing in creating new product features for the Anvilogic Detection Engineering Platform.
He began his career at T-Mobile by designing and implementing a company-wide customer scoring system to identify the optimal customers for new product offerings, such as wireless home internet.
Dillon then joined the Cyber Security team to work as ingestion lead for the Cyber Security Data Platform — a revolutionary “Big Data” security solution that ingests, normalizes, enriches, triages, and manages application and security data at scale in real-time.
Then, he went to AttackIQ to develop new product features and simulated attacks for their breach and attack simulation platform.
Experience
Data DevOps Engineer at Anvilogic focused on building new product features
– Architect, design, and develop new core product feature - Raw Data Pipeline:
- Leverage Python and AWS infrastructure (API Gateway, Lambda functions, S3, IAM) to automatically ingest customers’ raw data from multiple sources into Snowflake
- Empower customers to easily build powerful detections across multiple datasets
Security Engineer at AttackIQ focused on customer success
– Develop new product capabilities:
- Tagging system in Django to support easy model tagging
- Logging capability for nested scripts in Python
– Identified at risk customers by creating an extensive business intelligence dashboard in Redash to highlight product use and pain points - presented findings and dashboard at company all hands
– Created cyber attack simulations of advanced persistent threat actors through replicating attack behaviors targeting specific nation states
– Configured Splunk, Palo Alto Panorama, CrowdStrike, and other EDRs to detect malicious attacks and malware
Solving cybersecurity problems by implementing innovative machine-learning and data science techniques.
– Collaborate to create and maintain Cybersecurity Data Platform – revolutionary platform for logging and responding to security events:
- Reduce security event search time from weeks to minutes
- Lead log ingestion and create data flows using NiFi and MiNiFi
- Mentor coworkers and delegate ingestion tasks
- Implement machine learning to cluster events and identify threats
- Enrich existing data to make it better (feature engineering)
- Create monitoring using Grafana, Prometheus, and Pager Duty
- Create Slack chatbot to aid in monitoring
– Create AI log parser that automatically parses security event logs using machine learning and natural language processing
– Develop single pane of glass search which cuts incident response time by 50% by enhancing all security events with username, server/computer, and Active Directory domain information
– Enable 2500 IDS/IPS signatures by creating a vulnerability scan dashboard in Splunk to intelligently determine which signatures to deploy
Offering consumers a better home internet solution.
Create and automate company-wide customer scoring system using Alteryx and Python to rank over 40 million customers for new product offerings:
- Clean and combine data from multiple sources
- Use machine learning to determine customer locations
- Coordinate with other departments to determine most profitable target customers
NPPI All Star of the Month – November 2018
Skills
Machine Learning
XGBoost, Random Forest, Neural Networks, Naïve Bayes, Logistic Regression, Support Vector Machines (SVM), K-means clustering, Principal Component Analysis (PCA), Natural Language Processing (NLP), etc.
Big Data
Splunk, Spark, Kafka, Google Compute, Azure, Storm, NiFi, MiNiFi, Metron, Kibana, Elasticsearch
Programming Languages
Python, R, Java, SQL, Matlab/Octave, some Javascript
Data Analysis/Visualization
Tableau, Alteryx, Matplotlib, Pandas, Seaborn, ggplot2, MapInfo
Monitoring
Grafana, Prometheus DB, PagerDuty
Patents
U.S. Patent Application 16541121
Filed August 2019
Camp, Jat, Dousson, Sandhu, Escudero 2019. Customer Experience Scoring on Mobile Network Systems and Methods. U.S. Patent Application 16541121, filed August 2019. Patent Pending.
U.S. Patent Application 16529697
Filed August 2019
Camp, Jat, Dousson, Sandhu, Escudero 2019. Dominant Customer Locations Identification System and Methods. U.S. Patent Application 16529697, filed August 2019. Patent Pending.
Certifications
Machine Learning by Stanford • MySQL by Duke University • Python and R by DataCamp • Six Sigma White Belt • Six Sigma Yellow Belt in progress
Projects
Detecting Credit Card Fraud with Cloud Computing
I use Python on a Google Compute virtual machine to analyze anonymized credit card transactions and successfully detect fraud with an F1 score of 85.68 and an accuracy of 99.96% (compared to a baseline accuracy of 99.83% when predicting that no transactions are fraudulent).
Using natural language processing (NLP) in R to classify Indeed.com job descriptions as Data Scientist or Data Analyst.
Spam Classification Using a Support Vector Machine in Python and R
https://dilloncamp.com/projects/spam.htmlSpam Classification Using a Support Vector Machine in Python and R
Using a support vector machine (SVM) in Python and R to classify emails as spam or not.
Using K-means clustering in Python and R to reduce image size.
Image Compression Using Principal Component Analysis (PCA) in Python and R
https://dilloncamp.com/projects/pca.htmlUsing PCA to reduce the size of facial images in both Python and R.
Education
Baylor University
Master's Degree, Economics
2017
• Full Scholarship, Dean’s List
• GPA: 3.96 / 4.0
Baylor University
BBA, Finance
2014
• National Merit Scholar – Full Scholarship, Dean’s List, Beta Gamma Sigma
• GPA: 3.98 / 4.0