Adrian Hills (@AdaTheDev) header
Adrian Hills (@AdaTheDev) profile picture

Adrian Hills

@AdaTheDev

I write code. I delete code. I build stuff. Often found on the Refactor Tractor. .NET / Azure / AI / Data

 Hampshire, UK
adathedev.co.uk
Joined February 2009
  • Tweets7.4K
  • Following288
  • Followers409
334Media
Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)Photo by Adrian Hills (@AdaTheDev)
Hey @_Evan_Boyle - is sandbox support on the roadmap for Copilot CLI?
1
51
Biggest barrier I personally see to further adoption of GitHub Copilot is how far behind the experience and supported feature-set feels in Visual Studio compared to VSCode, Cursor etc. And combined with Copilot CLI being in Preview, it shifts attention elsewhere.
1
114
Yeah, I could switch out to VS Code for Copilot and then back to VS. But would be great to have better first class support in VS for the in-IDE experience. I'm increasingly switching out to Codex CLI, which is working very well. Keen to make use of Copilot CLI once it goes GA
1
43
One of my biggest UX gripes with GitHub Copilot (in VS) is that you have to wait for the whole response to be generated before you can read it due to the auto-scrolling as the response streams in. Would be a big plus to be able to stop the auto-scroll!
57
Anyone else having issues with Azure OpenAI (timing out trying to connect via API, failing to load model deployments in Azure OpenAI Studio...)?
6
3
289
Anyone else having issues getting into the Azure OpenAI Studio today?
4
171
Good article (Azure OpenAI) on considerations for right-sizing your Provisioned Throughput deployment, spilling peak workloads over to PAYG for cost optimisation: Right-size your PTU deployment and save big (microsoft.com)
1
84
Transitioning to Production with LLM capabilities, when you care about consistency of guaranteed throughput and latency is a key challenge that needs to be considered carefully up front.
1
1
67
PAYG models are great and have low outlay, but usually don't have guarantees and usually higher latency. Provisioned Throughput models give consistency and lower latency but a huge jump up in committed spend, where you end up paying for a lot of quiet time in a non 24/7 scenario
1
1
43
Full Fibre 900 - £41.99/month Full Fibre 900 for a customer of many years (me) - £55.54/month Don't know what's going on with @Plusnet these days, but that's shocking. Making me wonder if longer term, it'll be wound down (as part of consolidating Plusnet/BT/EE, per mobile)
1
133
Any of my followers (or beyond) have experience with Amazon Bedrock for hosting LLMs (on a token usage basis and/or Provisioned Throughput basis)? Keen to hear thoughts, experiences etc. Please RT!
1
64
Good to see a new Learning Path for Semantic Kernel on MS Learn: devblogs.microsoft.com/seman…

Learning Paths for Semantic Kernel | Microsoft Agent Framework

We are excited to highlight new learning content released focused on Semantic Kernel. APL-2005 Develop AI agents using Azure OpenAI and the Semantic

devblogs.microsoft.com
2
59
Switched over recently to the new gen Planners (Handlebars) as part of Semantic Kernel v1.x upgrade. Next step is to optimise the "plan generation" phase of a workflow. Generating plans is expensive (time and cost) and relatively prone to error.
1
1
70
To save generating plans afresh every time for similar inputs, I'm building a prototype Semantic Kernel "Plan Cache". Still needs LLM (+ Vector Search), but with a much smaller input context, and a more focused task.
1
1
47
Having significant latency issues with GPT 4-Turbo (1106-Preview) on Azure OpenAI, making it consistently, noticeably slower than GPT-4 (0613). Anyone else encountering this? e.g. 48.8s (GPT-4-Turbo) vs. 13.9s (GPT-4) Snapshot times, so fluctuates, but wildly so for 4-Turbo.
1
1
154
But the trend is pretty consistent on Azure OpenAI with 1106-Preview. Running tests on isolated model deployments, using low token usage test case (<500 prompt tokens, < 300 completion tokens), various points throughout the day, different machines/code implementations etc etc
1
72
Currently watching: status.dev.azure.com/_event/…
1
31
Microsoft Graph activity logs (learn.microsoft.com/en-us/gr…) is long overdue. Finally able to get a tenant-wide view on usage - dusting off my KQL skills!
Access Microsoft Graph activity logs for tenant monitoring - Microsoft Graph

Access Microsoft Graph activity logs for tenant monitoring - Microsoft Graph

Microsoft Graph activity logs provide a detailed audit trail of all API requests in your tenant, helping you monitor and investigate activities. Learn how to access and use these logs.

learn.microsoft.com
2
56
Finishing up prep for an internal talk I'll be giving tomorrow on the Semantic Kernel. There are demos, and then there are Generative-AI demos - let's hope it all behaves!
1
59
Enjoyed the keynote, as you'd expect!
Image from Adrian Hills (@AdaTheDev)
1
58
Checked in, ready for MS Build AI day to start
Image from Adrian Hills (@AdaTheDev)
3
154
Couple of days in the London office, and decided to take the scenic route to the hotel
Image from Adrian Hills (@AdaTheDev)
1
69
Just signed up for the Microsoft Build: AI Day at the London ExCel on 19th Oct - if anyone I know is heading there too, give me a shout :) msevents.microsoft.com/event…
1
2
239
I've been using the MS Semantic Kernel for a couple of months now to incorporate LLMs into POCs. At the heart of everything I do in this space now. Recent focus has been on observability, getting prompt & completion token usage metrics. Cost & throughput optimisation important
1
1
125

No more items