With the rate of change in AI today, the answers to "what is possible" and "what is the easiest way to do it" change on an almost daily basis. Courses from classic training providers go out of date quickly, and figuring out where to start and what tools to learn is hard and filled with distracting rabbit holes.
Because we are using these tools and techniques every day to deliver real products, we constantly stay up to date, and prioritise well supported and popular industry standard tools which allow us to deliver cutting edge models and products as easily as possible.
With a focus on practical and pragmatic end to end learning, our courses provide a solid foundation to developing and deploying cutting edge neural networks and machine learning using the most popular industry standard libraries, tools and services available.
Previously responsible for technical training at Deloitte, and constantly working to upskill the data scientists around him; Will has been writing and delivering technical training courses for over a decade.
Practical Deep Learning (1-3 days)
Using the Fastai and PyTorch deep learning stack, we run through a range of problems with hands-on training of models in a dedicated per-participant training environment. We cover modern computer vision, language and structured tabular examples and exercises. This course can be tailored around your use cases and data (see below), or extended to include deeper techniques around specific areas.
From Model to Product (1-2 days)
Even the most performant and powerful models cannot release value in isolation. In this lab-style training course we explore some examples of taking models from research to real products. From read-only BI tools such as Tableau / Power BI dashboards, prototype applications in streamlit , to serverless SPA applications delivered on the cloud. We will look at dev ops and the strengths and limitations of different services available today.
Statistics & Multiple Linear Regression for Data Science (1-2 days)
A review on the use of statistics in Data Science. Using hypothesis testing, A/B tests, and linear and logistic regression models to understand and test relationships in data. We'll talk about Z and T-testing, Non-parametric tests, model interpretation, p-values, effect size, sample sizes and power calculations.
We find that the best way to learn is invariably on the job. Whilst our practical training courses simulate this with real world examples and case studies, why not achieve two goals at once?
Combine our AI & ML Model Development or AI Product Design & Software Development services with on-the-job training sessions, in the form of workshops and hackathons, to help move your opportunity forward whilst taking your team with it.
Ultimately leading to sustainable solutions that are understood, maintainable and upgradable in-house, this provides a great hybrid option between outsourcing your data science and developing in-house talent.
Against the constant background noise of marketing and always-evolving jargon in data science, it would be easy to miss the step-change that has occurred with AI in the last 5 years.
There has been a fundamental change in the ease with which machines can be taught to read, write, see, hear and understand. Objectives that were once moonshots; the domain of university departments and large R&D teams, can now be comfortably dispatched by a well-motivated graduate with a few weeks training.
This occurred first in computer vision, but more recently has been brought to natural language processing; reading and writing plain English (or any language). The implications for business cannot be overstated. The wave of automation will change the economics of many existing products. At the same time, it will make many new business cases technically and economically feasible for the first time.
Due to the speed of this change, and the associated noise - there is a lot of misinformation - particularly as old limitations and caveats continue to get recited years after they are no longer true.
In this session, we look at the current state of AI; what is possible now; what is not, and what is likely to be possible very soon. We demystify AI models, and talk about how they differ from classic analytics and machine learning. We take a look at the economics of AI, the opportunities, and ways to understand the likely short and long term impacts on your industry and business.
Finally, we often follow these sessions with an AI Strategy workshop to to consider AI opportunities across your organisation, to ensure you have actionable opportunities to follow up on.
