Uncharted Software
Senior Software Architect
May 2016 to 2018 · Toronto, ON, Canada
Uncharted Software is a data visualization company with products used by government, law enforcement, financial industries.
Due to the sensitivity of work involved, certain details listed below have been obfuscated or redacted.
Highlights
- Partnered with data scientists and designers to build early prototypes of a causal-graph visualization tool, translating complex relationships into clear, interactive visuals used for model refinement and product design.
- Implemented Natural Language Processing (NLP) systems to perform Named Entity Recognition (NER) and extract features from unstructured and semi-structured data.
- Explored emerging NLP technologies such as spaCy and Word2Vec to model relationships between entities — foundational experience that later informed an intuitive understanding of modern LLM-based systems.
- Optimized large-scale data pipelines (Scala, Spark, HBase, Hadoop, Elasticsearch) to improve performance and information retrieval speed.
- Redesigned search schemas and algorithms to enhance result relevance and system responsiveness.
- Contributed to a culture of clarity through visualization, bridging the gap between engineering, data science, and design.
Projects
[redacted project], 2017-2018
Node.js + Express · RDF · Turtle (TTL) · SPARQL · D3 · Docker
- Built back-end framework of new product
- Built RDF/Turtle/SPARQL parser for causal graph traversing and visualization
- Built full-stack prototype of interactive causal graph
- Worked closely with data scientists to develop early prototypes of product
TellFinder, 2016-2017
Python w/ spaCy · Scala + Spark · HBase · Hadoop · Elasticsearch · RabbitMQ
- Optimized Elasticsearch model for improved query response time and searchability
- Used NLP and named entity recognition (NER) via spaCy to extract data from unstructured and semi-structured data
- Refactored legacy Java code to Scala with rich set of unit tests