ngrok的动态

?? ?? We just shipped a major redesign to the traces view. This is the number 1 requested change by most of Langtrace users. Sharing a preview of what it looks like. Some notable callouts: - Resizeable columns. - Hideable columns. - Sheet view for spans. - Hover over spans to see attributes, events/errors. - Click into spans to get a dedicated view of attributes, events/errors and conversations. - User preferences are automatically saved. More to follow ??

Checkout new Design and Feature Updates on LangTrace.ai #llmops

Encountered a crash related to Compose and ViewModel. Quick search led me to this issue. Googlers have known about this critical bug since Jan 2023, and, at some point, marked it as "closed", even though crash reports kept coming. Now this bug made it into "stable" tools. Shows the risks associated with adopting new tech, like Compose: it can take years, literally, for these tools to mature and become fully reliable. https://lnkd.in/dchvJvWM

Time : O(n) Space: O(n) approach : start pushing the nodes of linkedlist on to a stack if and only if top of the stack is greater else, pop peek of the stack and push it

Heard about the Speculation Rules API? You can use it to prerender future navigations for instant future navigations. Well, did you know that Chrome prerenders pages from the address bar and the bookmarks bar? So your site might already be using this prerendering technology. Use the "Navigation Type Distribution" tab of the CrUX Dashboard to check your site to see if, and how often, that happens: https://lnkd.in/ekCzZFrB

In this GeoTalks Express, learn about the new features in Global Mapper v25.1 while our experts demonstrate them: https://ow.ly/9shn50QJRaT

?? LoRAX v0.9 is out! By number of contributions, this is by far our biggest release to date. Excited to see the community continue to grow! What's new: ?? Overhauled LoRA memory manager. No more CUDA OOMs, LoRAX will now automatically swap and page adapters when it hits the memory limit of its resource pool. ?? New model architecture: Qwen 2! ?? Optional max_new_tokens: by default, LoRAX will now generate up to the max_total_tokens, so no more need to mess with this param in most cases. https://lnkd.in/e6XbRKs4

?? Scaling High-Performance Caches at DV In the world of in-memory caches, ChronicleMap has emerged as a game-changer for scaling high-performance systems. DoubleVerify (DV) faced the challenge of maintaining low-latency, high-throughput in their pre-bid verification solution while managing massive datasets. After evaluating several caching solutions, DV chose ChronicleMap for its low garbage collection overhead, off-heap scalability, and seamless integration into existing infrastructure. ChronicleMap not only improved cache performance but also reduced code complexity and startup times, making it the ideal solution for DV’s growing data needs. Read the full blog here: https://lnkd.in/di6GyWTv #ChronicleSoftware #TechInnovation #PerformanceOptimisation #ChronicleMap #InMemoryCaching

Fully self-healing code might not be possible yet, but with the BEAM and LLMs you can create something similar: ?? https://loom.ly/Dki-7-Q #ElixirLang #MyElixirStatus

?? Traverxec Machine Conquered! ???? Thrilled to share that I’ve completed the "Traverxec" machine on Hack The Box! This challenge pushed my skills in web server exploitation, enumeration, and privilege escalation. Key takeaways: 1?? Used tools like nmap for in-depth scanning. 2?? Exploited nostromo v1.9.6 in order to get remote code execution. 3?? Exploited journalctl with less in order to escalate privileges vertically. writeup :--> https://lnkd.in/ggpHAZZr try it :--> https://lnkd.in/gWiBUjiB