Dasith Wijes
@dasithsDeveloper, blogger, tech speaker with unhealthy focus on distributed systems, overly keen problem solver and improving cricketer. Snr Engineer @Microsoft.
Similar User
@_AndrewB
@DDDAdelaide
@yashints
@todthomson
@staff0rd
@BenW_Pearson
@xwipeoutx
Had an awesome time today at #apidays and met a lot of new faces and some familiar ones as well. The slides for our talk about software supply chain security are here speakerdeck.com/dasiths/buildi… @NotaryProject @OpenPolicyAgent @orasproject @AquaTrivy @GrypeProject
My team have been working with #PromptFlow recently and we did some benchmarking of pf-serve to find throughput problems. I wrote up a blog post with our findings here. devblogs.microsoft.com/ise/promptflow…
Ah @stephentoub’s annual .NET performance post is here! devblogs.microsoft.com/dotnet/perform… #dotnet
My team recently concluded a project that used #promptflow. There were some learning around optimizing for throughput and latency. We developed an awesome little throughput testing kit for pf-serve and contributed it to the Promptflow OSS repo. github.com/microsoft/prom…
Highly requested docker file feature coming in the next version of aspire github.com/dotnet/aspire/… #dotnet #aspire
I wrote about a simple pattern you can follow when developing #LLM apps to prevent a certain class #prompt injection attacks with regards to user enumeration when using tools. Examples using #langchain. dasith.me/2024/05/03/llm…
As long as AI systems are trained to reproduce human-generated data (e.g. text) and have no search/planning/reasoning capability, performance will saturate below or around human level. Furthermore, the amount of trials needed to reach that level will be far larger than the…
Interesting how in all these domains AI is asymptoting at roughly human performance - where's the AI zooming past us to superintelligence that Kurzweil etc. predicted/feared?
Legit question. What is the value of frameworks like LangChain, Autogen, crewAI, ... that basically build the same abstractions on top of a programming language that the underlying programming language already supports. For example chaining is sequential composition, agents in…
Wait… what!? Fault tolerant 2PC that is simple and commits in 1RTT? 🤯 If you missed @ChrisJe34211511 fantastic talk in #eurosys24 (#papoc24) definitely check the paper bit.ly/U2PCp
Most application developers likely dont understand ACID transactions well, but I wonder how many believe transactions are important for preventing bugs anyway. What do you think? blog.acolyer.org/2015/09/08/out…
🚨UPDATE your @Apple products now! We @citizenlab w/TAG's @maddiestone caught #predator spyware attacks against a prominent pro-democracy Egyptian politician after he announced presidential ambitions. Apple rushed a patch. It gets crazier 1/ citizenlab.ca/2023/09/predat…
Have been experimenting with "plan and execute" agents for extending capabilities of LLM & GPT4 to consume "live" data sources. Making some awesome progress and cool demos. Check it out. I call it a "knowledge mesh" #llm #langchain #gpt4 github.com/dasiths/llm-pl…
Container registries are no longer a place just for storing container images. You now have HELM packages, supply chain artefacts etc that need the same home as your container images. Enter OCI 1.1 spec and ORAS. #oci #oras #k8s #notation dasith.me/2023/06/04/wha…
When we interact with language model APIs, such as the OpenAI API, we typically have to set a “temperature” parameter when obtaining output from the language model. But, what exactly is this parameter and how does it work? Let’s take a deeper look… The decoding process:…
A new version of #MLdotNET Model Builder is now available. AND, it includes support for object detection using your local CPU or GPU. Read about the updates we made that enables integration of custom machine learning models into your #dotNET apps. 📖 msft.it/6019gbKyf
So many parallels youtu.be/n9cdGXa-uyM
1/The call for a 6 month moratorium on making AI progress beyond GPT-4 is a terrible idea. I'm seeing many new applications in education, healthcare, food, ... that'll help many people. Improving GPT-4 will help. Lets balance the huge value AI is creating vs. realistic risks.
Solution: - Write a test in your build pipeline that checks and validates your *production* certs - Fail your builds when they're within ~6 weeks of expiry - Certs break so many people this is literally as important as your ability to ship
Lol. iPlayer is down. Wife asked what the problem was likely to be. I said "DNS or certs" - sure enough, the TLS cert expired at 3pm today. Always one of those two troublemakers!
United States Trends
- 1. Georgia 213 B posts
- 2. Haynes King 8.816 posts
- 3. Kirby 22,6 B posts
- 4. Iowa 23,6 B posts
- 5. Nebraska 14,9 B posts
- 6. BOB BRYAR 5.320 posts
- 7. Carson Beck 3.302 posts
- 8. GA Tech 2.947 posts
- 9. Athens 6.887 posts
- 10. 5th OT N/A
- 11. #SmackDown 55,1 B posts
- 12. Targeting 76,3 B posts
- 13. #GTvsUGA N/A
- 14. 3rd OT N/A
- 15. Dawgs 8.769 posts
- 16. Brent Key N/A
- 17. Chiefs 167 B posts
- 18. Joe Tessitore N/A
- 19. The SEC 248 B posts
- 20. #OPLive 3.681 posts
Something went wrong.
Something went wrong.