AI

Strategies for Addressing Tech Debt

Ignoring tech debt because it’s expensive doesn’t make the problem go away — it only kicks the can down the road. Businesses should approach technical debt as a routine, scheduled part of their development process.

From RAGs to Riches – Adding Context to Your LLM

In my previous post, Experiences in Fine-Tuning LLMs: Time + Power = Potato?, I covered my experiences around trying to fine-tune an LLM (large language model) with a dataset, which gave me less than stellar results. Ultimately, fine-tuning is best for a use-case where additional reasoning & logic needs to be added to an LLM, … Read More

Experiences in Fine-Tuning LLMs: Time + Power = Potato?

Embarking on the journey to fine-tune large language models (LLMs) can often feel like setting sail into uncharted waters, armed with hope and a map of best practices. Yet, despite meticulous planning and execution, the quest for improved performance doesn’t always lead to the treasure trove of success one might anticipate. And I know you … Read More

Philly ETE 2019 – Leemay Nassery – Neural Networks IRL (In Real Life)

Abstract What exactly is a Neural Network? How do you go from experimenting with a deep learning model in the notebook of your choice to deploying your model to production? What is the secret sauce to a successful feature or platform that utilizes “AI” to accomplish the business goals or needs of your platform? And … Read More

Philly ETE 2019 – Anatoly Polinsky – Machine Learning: from ABCs to DEFs

Abstract I’d like to introduce you to this new, 60 year old, kid on the block: “Machine Learning”. Some math + some stats, but mostly “what”s, “why”s and “how”s of different problems it solves, and of course some code, since that’s what machines speak best. While we’ll ride along with mouthfuls such as “stochastic gradient … Read More

Philly ETE 2015 #26 – Soumith Chintala – The Deep Learning Revolution: Rethinking Machine Learning Pipelines

In the last decade, a class of machine learning algorithms popularly know as “deep learning” have produced state-of-the-art results on a wide variety of domains, including image recognition, speech recognition, natural language processing, genome sequencing, and financial data among others. What is deep learning? Why has it become so popular so quickly? How can one fit deep learning into existing pipelines?