A rolling guide on how to push your company's frontier.


Back Propagation: aka The Witchcraft of AI and Pytorch

This post is for readers interested to know the wizardry of AI behind the curtain. It is an attempt to explain the principle of learning via back propagation, all the way down to computation and in the least burdensome manner possible.


The sorcery of fine-tuning: unleash magic from enterprise data.

Enterprises will increasingly leverage large language models (LLMs) in order to gain a competitive edge. Fine-tuning is essential for success because it enables LLMs to unlock high-value use cases with existing data.


The New Innovation: Right-brained AI

GenAI is still in its infancy, yet many leaders already want to know what they should do differently beyond exploring tactical use-cases.

Our Head of AI, Paul Golding offers a more strategic outlook through a familiar lens: innovation.


Do LLMs really understand prompts? What do they grok?

This post gives a dummies guide to natural-language “understanding”, especially the special use of the term “understanding”. Do Large Language Models (LLMs) like ChatGPT actually grok our prompts? The answer is yes, and no. We explain why grasping the difference is important.


Unlocking the AI Bounty with Direct AI

Whilst ChatGPT has fired the imagination, hundreds of AI advances are made monthly. Think of them as keys waiting to unlock hidden innovation in product, marketing and operations. We call this unlocking the AI bounty. A scalable way to unlock the bounty is to enable as many workers as possible to apply AI directly to their work, as easily as using a spreadsheet.

Let's build your AI frontier.

The field of AI is accelerating. Doing nothing is going backward. Book a 1:1 consultation today.