Measuring LLMs with Jodie Burchell
.NET Rocks!3 Huhti

Measuring LLMs with Jodie Burchell

How do you measure the quality of a large language model? Carl and Richard talk to Dr. Jodie Burchell about her work measuring large language models for accuracy, reliability, and consistency. Jodie talks about the variety of benchmarks that exist for LLMs and the problems they have. A broader conversation about quality digs into the idea that LLMs should be targeted to the particular topic area they are being used for - often, smaller is better! Building a good test suite for your LLM is challenging but can increase your confidence that the tool will work as expected.

Jaksot(1961)

Jon Harrop Makes Us F#

Jon Harrop Makes Us F#

Jon Harrop introduces Carl and Richard to F#, a functional language that runs under the CLR. F# performs like C#, but being a functional language, has interactive scripting (similar to Python) but is rooted in the strong type inference and safety that other functional languages like ML focus on. Being in the CLR means you can build certain parts of your application in F# and then reference them from other languages, the same way VB.NET and C# interoperate.Support this podcast at — https://redcircle.com/net-rocks/donations

23 Elo 20071h 4min

Remi Caron Develops with Off-the-Shelf Software

Remi Caron Develops with Off-the-Shelf Software

Remi Caron, one of the organizers of the SDC conference in the Netherlands, tells Richard and Carl how using standard toolsets and software packages helps him stay focused on his customers' software problems and deliver more powerful solutions faster.Support this podcast at — https://redcircle.com/net-rocks/donations

21 Elo 20071h 15min

Donald Farmer on Data Mining

Donald Farmer on Data Mining

Donald Farmer talks about data mining with SQL Server and related technologies, including a fascinating discussion about using algorithms for predicting future trends.Support this podcast at — https://redcircle.com/net-rocks/donations

16 Elo 20071h 1min

Udi Dahan talks SOA Sense

Udi Dahan talks SOA Sense

Udi Dahan calls in from Israel to talk common sense about SOA. His pragmatic approach to the topic is refreshing and timely.Support this podcast at — https://redcircle.com/net-rocks/donations

14 Elo 20071h 1min

David Hayden on the Enterprise Library

David Hayden on the Enterprise Library

Carl and Richard talk to David Hayden about the new features of the Microsoft Enterprise Library 3Support this podcast at — https://redcircle.com/net-rocks/donations

9 Elo 20071h 6min

Phil Haack on Subtext and Open Source

Phil Haack on Subtext and Open Source

Carl and Richard talk to Phil Haack about his work with Subtext (a derivative of the .Text blog software package) and his work on various open source projects.Support this podcast at — https://redcircle.com/net-rocks/donations

7 Elo 20071h 11min

Sandcastle!

Sandcastle!

Carl and Richard talk to Anand Raman and David Wright about Sandcastle, an internal tool for generating code documentation that is now available to the general public.Support this podcast at — https://redcircle.com/net-rocks/donations

2 Elo 20071h 6min

Dan Ciruli's Grid Computing Redux

Dan Ciruli's Grid Computing Redux

Dan Ciruli of Digipede Technologies is back to bring us up to date with Digipede Networks, a .NET toolset for enabling grid computing.Support this podcast at — https://redcircle.com/net-rocks/donations

31 Heinä 20071h 7min