Rayne K. Wilde

Primarily responsible for developing data science solutions for analysts and the DoD. Responsibilities vary from project to project and depend on customer needs. Deploying containerized apps on protected networks and embedded systems.

• Unsupervised Anomaly Detection: Developed an ML model to identify anomalies in advance, alerting analysts to potential issues
• Dockerized Applications: Developed Multiple Frontend and backend containerized applications to support the intelligence community
• Text Extraction Location: NLP techniques to extract locations in nested patterns from general to specific and apply geocoordinates.
• Knowledge Graph Optimization: Enhanced scalability, search relevance, and reduced latency for large datasets.
• Generative AI Search: Integrated phonetic matching, synonym handling, and semantic search within vector databases.
• Hybrid Search System: Engineered a custom ElasticSearch alternative using FAISS, Weaviate, and ANN.
• Automated Vulnerability Scanner: Developed a YARA scanner to detect malicious code using CVE and MITRE ATT&CK standards.

Lead Machine Learning Engineer for a data science team. Tasked with providing language-agnostic NLP-specific insights, analysis, and data handling.

• Stance and Sentiment Detection: Developed models to classify tone and stance, with analysis pipelines for sentiment insights.
• Data Compression and Deduplication: Improved storage efficiency by streamlining datasets with data cleaning and clustering.
• Data Analysis Platform: Created a scalable library for document tagging, grouping, and clustering to support various analysis goals.
• NLP ETL Pipeline: Developed a language-independent pipeline for data extraction, normalization, multi-stage compression, deduplication, and embedding preprocessing.

Intelligence production and mentorship.

• Mentored 3-5 airmen in creating intelligence products.
• Creating, Reviewing, Disseminating quality intelligence products.

Staffing Agency Contract to Hire for Occam Solutions

• Automated intelligence analysis, increasing speed, reliability, and data accessibility through web-based tools tailored to client specifications.
• Created document classification models with ANNs and semi-supervised learning, along with a multi-stage pipeline for document categorization.
• Document Classification: Implemented ANN-based classification for categorizing text, with a pipeline for tagging and cluster analysis.
• Multi-Stage Data Aggregation: Built a document processing workflow for classification, summarization, and cluster analysis, enhancing document insights.

Focused on learning QIR, IIR, LLVM, Compiler Basics, and Graph Theory

• Worked on innovating and optimizing routing within a Compiler
• Developed the foundation for a sandbox to allow for user simulation
• Focused on Ion Routing, Simulating Routing in the Compiler
• Implement Multiple Graph Algorithms to Optimize Routing

Data Science Professeional performing a range of Data Analytics, and EDA tasks

• Created Causal and exploratory reports based on App Data
• Responsible for Data Cleaning, Exploratory Data Analysis (EDA), Visualization, and modeling

Intelligence production and mentorship.

• Created a Machine Learning program to correlate Text to Corporate metric success probability
• Classified information varying in both location and type
• Performed Exploratory and Causal quantitative analysis, as well as performed subjective analysis
• Text Analytics; Created a program to associate work logs to probability of project success, From an SQL server, used unique words in work logs
• Causal Analysis: Created multiple reports from corporate metrics to project success