AI in Production

Join us for the first installment of the Jumping Rivers AI in Production conference!

Hosted in the centre of Newcastle Upon Tyne, UK, this conference will delve into the world of AI and Machine Learning.

Find us at The Catalyst, 3 Science Square, Newcastle Helix, Newcastle, NE4 5TG.

Want to support our conference? Get in touch!

Abstract submission will open soon.

Abstract submission is now closed.

AI in Production will take place over two days, split into a day of workshops and a day of conference talks. Tickets are available for either the conference only, or for the conference and one of the workshops.

We will also be hosting a dinner reception on the evening, which is included in the price of your ticket!

Day 1 will consist of workshops run by some of our very own Jumping Rivers pros and other experts in the field. Particpants can take two half-days workshops and lunch is included.

Morning Workshops ( – )

• Prompt Craft & AI Integration: Building LLM-Driven Workflows in R and Python

  Myles Mitchell (Jumping Rivers)

  This workshop will provide an introduction to Large Language Models (LLMs) and their applications in Data Science. We will start by exploring how LLMs work, how they are priced and their strengths and weaknesses. We will then learn how to work with LLMs programmatically, including:

  • Sending user prompts to an LLM API using the ellmer and chatlas packages.
  • Including additional instructions via the system prompt.
  • Formatting your prompts to receive a structured, tabular output.
  • Using LLMs to summarise images and PDFs.
• Nothing to Gold… Productionising with Databricks using the Medallion Architecture

  Declan Watson and Liam Wilkinson (Databricks)

  A hands-on look at how teams can turn messy, scattered data into clean, trusted, production-ready assets. Expect practical tips, real patterns, and the kind of “oh, that’s how it works” moments that make the Medallion Architecture click. A great session for anyone wanting to move faster from idea to usable data product.

• Improving your workflow with Positron and AI

  Kia Mack (Jumping Rivers)

  A demonstration-led workshop introducing participants to Positron and its AI powered capabilities through the Positron Assistant and DataBot. By the end of the session, participants will be comfortable navigating Positron, using AI assistance effectively in their coding and data exploration workflows, critically evaluating AI-generated code and insights, and creating custom agents and prompts for more helpful, efficient and bespoke assistance. This workshop will primarily use R for demonstrations, but all concepts covered will be applicable to both R and Python workflows in Positron.

Afternoon Workshops ( – )

• Shiny Meets LLMs: Smarter App Experiences

  Myles Mitchell (Jumping Rivers)

  In this hands-on workshop we will discover different ways of harnessing the power of Large Language Models (LLMs) in Shiny applications, including:

  • Connecting to a model of your choice using the ellmer and chatlas packages.
  • Inserting a chat box in your Shiny application using the shinychat package.
  • AI-driven data exploration with querychat.
  • Setting up a retrieval augmented generation (RAG) workflow for incorporating domain-specific knowledge.
  • Safety, security and responsible use of AI.

  We will provide code examples in both R and Python.

• The Power of Databricks Genie Rooms… Data Discovery and Questions with Minimal Effort

  Declan Watson and Liam Wilkinson (Databricks)

  This session shows how Genie Rooms remove the friction from finding answers. Ask questions naturally, explore data effortlessly, and uncover insights without the usual back-and-forth or technical overhead. Perfect for anyone looking to boost self-service and make AI-assisted discovery a reality in their organisation.

• Self-hosted LLMs: Running Your Own Inference Infrastructure

  Daniel Burkhardt Cerigo (datavaluepeople)

  In this workshop we'll use popular open-source inference frameworks to set up your own LLM inference endpoint. Not every AI use case belongs on a third-party API like OpenAI; data sovereignty, compliance, or simply cost at scale can push you toward self-hosting. We'll cover when self-hosting makes sense, get hands-on deploying a working system, and then explore monitoring the metrics that matter in production - time to first token, tokens per second - and how to tune for your workload.

Dinner Reception ( - )

All tickets include entry to the dinner and drinks reception on night. Hosted in the atrium of the Catalyst building, this will be a great opportunity for networking and relaxing before the following day of talks!

Day 2 will consist of a range of talks from experts across industries. Registration begins at , followed by the first talk session at . The final session is scheduled to conclude around .

This year's conference features two streams:

Engineering Stream

Sessions focused on building, shipping, maintaining, and scaling AI systems and data pipelines. This stream will suit people working with infrastructure, deployment, monitoring, and production readiness.

Machine Learning Stream

Talks covering model development, evaluation, responsible use of data, and lessons from real projects. This stream will suit people who work with modelling, research, and applied ML

Time	Speaker

Situated just a short walk from the iconic St James' Park, The Catalyst is home to both the National Innovation Centre for Ageing and the National Innovation Centre for Data. It forms part of the Newcastle Helix, the state-of-the art innovation district in the centre of the city.

The Catalyst is just a ten-minute walk from Newcastle Central Station, on the East Coast mainline, with regular trains to and from Edinburgh and London taking around ninety minutes and three hours, respectively.

The closest airport is Newcastle International, which is approximately 30 minutes on the Metro from Newcastle Central Station.

AI in Production will follow the code of conduct as laid out by the R Consortium. Contact events@jumpingrivers.com if you have any concerns or wish to report an incident.