Skip to content
View connorcrowe's full-sized avatar
🗺️
🗺️
6 followers · 6 following

Block or report connorcrowe

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
connorcrowe/README.md

🌎 Data Background, Product Focus, Geospatial Passion

Hi, I'm Connor--- a data-driven product manager with a background in computer engineering & data science and a passion for geospatial problems. I've built products in deep tech, health-tech and education-tech.

I am on a journey in the geospatial industry to leverage my skills towards the problems that I'm most personally passionate about - climate change, urban development, & transportation.

Featured Projects

🚧 Current WIP: Geoff, the Spatial-AI Map-Maker

GEOFF (GEOspatial Fact Finder) aims to turn natural language prompts like "how many bike lanes are near school zones?", turn them into Spatial SQL, and display the results on a web map. Project progressing ~ 🔗 GitHub Repo

🗺️ High-Resolution Automated LULC Classification at Scale in Toronto

Using trained LULC U-net classifier to autoamtically predict land use land cover of the entire City of Toronto aerial at 2 px per meter. Technologies: gdal, leaflet, Vite, Tensorflow/Keras, Python

🗺️ Live Demo | 🔗 GitHub Repo

🛩️ Land Use Land Cover Neural Network Classification from Aerial Data

Trained a Convolutional Neural Network (CNN) with U-net architecture on aerial imagery to classify land use in downtown Toronto. Technologies: CNN, U-net, Tensorflow/Keras, Map Digitization

🔗 GitHub Repo

🌆 Urban Heat Island & Vulnerability Analysis of Toronto from Satellite Imagery

Mapped urban heat islands in Toronto using remote imagery and overlaid demographic data to highlight vulnerable communities. Technologies: QGIS, GDAL, Raster Analysis, Remote Sensing, Landsat

🗺️ Full Story on StoryMaps! | 🔗 GitHub Repo

🚲 Geospatial Analysis of Toronto Bike Share Data

Analyzed Toronto's bike share data (2016-2024) using spatial SQL and geospatial visualization to assess impact of changes in infrastructure. Technologies: PostGIS, QGIS, Python, PyQGIS

🗺️ Full Story on StoryMaps! | 🔗 GitHub Repo

Skills, Tools, Learning

  • Product Management
    • Prioritization, organisation, storytelling & communication
    • Agile, Jira, Git, etc.
  • Data & Programming
    • Python, Data, Machine Learning (Pandas, Geopandas, PyQGIS, Tensorflow/Keras, scikit-learn)
    • SQL (MySQL, PostGIS, BigQuery)
  • Courses
    • Fundamentals of Remote Sensing and Geospatial Analysis, Udemy, Matt Thompson
    • Machine Learning in GIS: Theory and Practice, Udemy, Kate Alison

Let's Connect

I'm always open to discussions about geospatial data, urban analytics, and climate tech.
💼 LinkedIn

Pinned Loading

  1. to-lulc-scale to-lulc-scale Public

    This project uses a U-Net CNN to classify land use for the entire City ot Toronto at high-resolution in an automated pipeline.

    Jupyter Notebook 8

  2. to-lulc-aiml to-lulc-aiml Public

    Land use land cover (lulc) classification of aerial imagery using machine learning techniques including U-Net architecture Convolutional Neural Networks (CNNs).

    Jupyter Notebook 1 1

  3. to-urban-heat-island to-urban-heat-island Public

    Geospatial analysis of remote imagery to identify where the urban heat island effect is worst in Toronto, Canada, and which areas have the population most vulnerable to them.

    1

  4. to-bike-analysis to-bike-analysis Public

    Statistical and geospatial analysis of Toronto Bike Share data and what it can tell us about the impact of changes to Toronto's bicycle infrastructure

    Jupyter Notebook 1

  5. dcp-sort dcp-sort Public

    A sorting algorithm designed for improved time complexity on a massively parallel system

    JavaScript 1