Jim Bennett
Principal Developer Advocate at Galileo
Redmond, Washington, United States
Actions
Jim is the worlds most energetic dev rel, and a Principal Developer Advocate at Galileo, focusing on enabling AI developers to be more productive by monitoring and evaluating LLMs and AI agents. He’s British, so sounds way smarter than he actually is, and lives in the Pacific North West of the USA. In the past he’s lived in 4 continents working as a developer in the mobile, desktop, and scientific space. He's spoken at conferences and events all around the globe, organised meetup groups and communities, and written books on mobile development and IoT. He is currently a Microsoft MVP for AI and Developer Tools.
He also hates and is allergic to cats, but has a 12-year-old who loves cats, so he has 2 cats.
Area of Expertise
Topics
Use an LLM 'off the grid' with local models
We've all got used to using LLMs in our developer workflow - from asking ChatGPT what tools and libraries to use, to getting GitHub copilot to generate code for us. Great when you are online, but not so useful when you are offline, like on the London Underground, in a plane with no WiFi, or in the middle of nowhere. But what if there was another way?
In this session, Jim will introduce offline LLMs using SLMs - small language models. We'll look at how you can run LLMs locally, such as Phi-3.5 from Microsoft, and add these to your developer workflow. We'll compare the performance of offline vs online, both the speed and quality, but also touch on privacy and other considerations. We'll also look at hardware requirements as we don't all have the latest GPUs to hand, showing how these models can run not only on very powerful laptops, but also on small devices like a Raspberry Pi for an LLM on your home network.
By the end of this session you will have an understanding of the technical differences between small language models and large language models, see how you can use them, and understand their advantages and limitations.
The force is strong in LLMs - building an open source Star Wars inspired copilot
In the Star Wars universe, many pilots have an astromech copilot. Luke had R2-D2 in the back of his X-Wing for example. As developers, we too have copilots. Although these are not as cool as R2, and don’t help us blow up the Death Star, they do help us with our day to day tasks like writing code.
Whilst copilots can be boring, Jim thought it would be fun to create one inspired by Star Wars to help him with important tasks, such as describing his Lego collection (Star Wars Lego of course), and helping him write code, all done in the style of a Jedi.
In this session, Jim will walk you through the steps to build your own copilot, using Microsoft.Extensiona.AI. By leveraging this SDK, your copilot can access an LLM of your choice from a range of cloud and on-device models such as Microsoft Phi and OpenAI GPT-4o. He'll also show the power of the MCP side, adding tools to research on Wookiepedia, or look at the scripts of the movies. And best of all, reply like Yoda, it can!
By the end of this session, you will be able to complete your apprenticeship and build your own open source AI copilot.
Prove going for a walk makes you more productive by measuring CO2 with an IoT device.
We've all been in meetings where we feel like our ability to process information is dropping by the minute. We get tired, our decision making ability drops, and often folks end up making terrible decisions. It turns out there is a simple reason, and it's the same reason our planet is heating up - Carbon Dioxide, CO2.
This is a simple problem to solve - when CO2 levels get too high, it's time to leave the room and ventilate. High levels of CO2 also imply stale air, and correlate with the spread of airborne diseases like COVID. The better the ventilation, the less CO2, and the less viruses in the air.
As developers, how can we solve this problem? We can do it by measuring CO2 levels with an Internet of Things, or IoT, device! This session will be hands on building out a CO2 alarm that will notify us if the level gets too high.
We'll start by introducing IoT and talk about the use cases from maker devices to industrial automation. Then will get going with the hardware, using a relatively cheap IoT device and a CO2 sensor. We'll then write code on the device in Python to gather the CO2 levels and send to the cloud. From there we can use the cloud to send an email once the CO2 level gets too high, alerting us to leave the room.
By the end of this session you will have an understanding of IoT and how to use it, along with knowing how to measure sensor data, send it to the cloud, and respond to events. You'll also have a new found appreciation for going outside between meetings.
NDC London 2025 Sessionize Event
KCDC 2019 Sessionize Event
Techorama Belgium 2019 Sessionize Event
NDC London 2019 Sessionize Event
Techorama NL 2018 Sessionize Event
NDC Sydney 2017 Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top