Articles
Rethinking Vector Search Experiences
Demonstrating UI components to deliver interesting vector search experiences.
Contemplation - Introspection in Python
Building a Python package to unlock the power of introspection.
Creating a Multiprocessing Context in Python
Implementing a Python package that uses introspection to automatically parallelise your code.
After All is Said and Indexed - Unlocking Information in Recorded Speech
Creating an end-to-end system for processing speech data, indexing it into a vector store and then using the information as context for a language model to provide a conversational interface.
Training a Neural Network with a Genetic Algorithm
Training a neural network to balance a pole on a cart using a genetic algorithm instead of backpropagation!
Deep Dreams in PyTorch
An implementation of Deep Dreaming using the PyTorch deep learning framework with a pretrained VGG19 convolutional neural network.
ConvNet Autoencoder for Scan Cleaning
Using a convolutional autoencoder to de-noise low quality and damaged scans of PDF documents with synthetic data for training, this is then applied to some real world examples of different types of issues with scanned documents.