Scala Center’s Post

✨ I'm delighted to share that I have completed my Google Summer of Code 17 weeks contribution with Ceph. 🔍 What I Did: - The Ceph project has a massive codebase, and maintaining storage reliability, performance, and resource efficiency is critical. - My project focused on identifying and fixing suboptimal code using clang-tidy, a powerful static analysis tool and integrating clang-tidy into Ceph's CI/CD. This was done in two phases: 🛠 Phase 1: Finding and fixing critical issues in the Ceph codebase using clang-tidy. 🔄 Phase 2: Creating a Jenkins pipeline for Ceph's CI/CD to analyze pull requests for potential code issues. 🌟 Key Highlights: ✅ Configured and finalized checks for clang-tidy. 🚫 Annotated false positives in the codebase to improve future analyses. 📋 Opened PRs to address identified issues. 🔍 Implemented the use-after-move check in Jenkins for pull requests. ⚙️ Developed and tested a Jenkins pipeline script to run clang-tidy checks and report issues in XML format. 🙏 I'm extremely thankful to my mentors Yuval Lifshitz and Ronen Friedman for guiding me throughout this journey. 🚀 Looking forward to continuing my contributions to Ceph and the open-source community! 🌐 (Note: Link of the work I have done is available in the comment section) #GoogleSummerOfCode #Google #GSoC2024 #LLVM #Clang #Jenkins

#Jaeger, the Distributed Tracing open source, is heading towards 𝗩𝟮 𝗺𝗮𝗷𝗼𝗿 𝗿𝗲𝗹𝗲𝗮𝘀𝗲! and the major theme is alignment with #OpenTelemetry 🔭 Jaeger is a graduated project under the Cloud Native Computing Foundation (CNCF), originally started as an internal tool at Uber. As the OpenTelemetry project emerged, there was clear overlap with Jaeger, and while certain steps were later taken (like deprecating the tracers client libraries in favor of OTel's), the need to maintain backward compatibility of the config, and need to transform to Jaeger's data model spec, has made it difficult to fully mitigate. This is where V2 comes in, with a re-architecture based on the OpenTelemetry Collector, with a promise of a simplified codebase, binary, configuration and more. This is the cloud-native open source community at its best, seeing the collaboration between CNCF's projects for unified open observability. And also seeing the great contribution of interns through the LFX Mentorship and Google Summer of Code programs, which brings new blood to an established project. 👏 A big round of applause to all the contributors to this major release! Read this excellent blog post by Jaeger creator Yuri Shkuro and maintainer Jonah Kowall on V2, the current Alpha version and roadmap to GA. https://lnkd.in/dWSYKwYb

🌟 Absolutely elated to announce my new Google Developer Student Club Certification! 🌟 Grateful to DHANA VEERA CHAITANYA KETHA and DEVI PRADEEP VASALA for the guidance. My special thanks to Radhika Mani Ma'am for introducing me to this program. #gcsj #googlecloudstudyjam #googlecloudcertified #googlecloudplatform

Day 34 of my Full Stack journey through Apna College's DELTA Feb 24 course was packed with valuable insights! Today, I delved into the following key concepts: 🔷 Understanding REST: Grasped the principles of Representational State Transfer (REST) and its importance in modern web development. 🔷 CRUD Operations: Learned the fundamental Create, Read, Update, and Delete operations within RESTful APIs. 🔷 Creating RESTful APIs: Explored the process of building robust backend systems from scratch. 🔷 Essential Routes: Mastered Index, Create, New, Show, Update, Edit, and Destroy routes, crucial for effective API navigation and resource manipulation. Excited to implement these learnings in my upcoming projects! 💡 #FullStack #WebDevelopment #RESTfulAPIs #LearningJourney 🚀

Consistently generating structured output from LLMs is challenging. Enter Outlines. Outlines enables you to generate structured content from LLMs such as JSON. You can achieve this in vLLM by: 1. Starting a vLLM server 2. Defining the desired structure 3. Passing the structure to the LLM Check out this notebook https://lnkd.in/dGSJDAxK to learn more. Join us on Slack for any questions https://lnkd.in/dbBgeYH3

🌟 Exciting News! 🌟 I'm thrilled to announce the release of Celest v1—a milestone our team has poured months of effort and innovation into. This release brings incredible new features: 💪 Introducing Celest Data - Celest now supports a new database layer built on #SQLite and the Turso platform, making database setup effortless with just a few lines in your Celest project! 🔒 Enhanced Celest Auth - Our home-grown auth solution is now completely open-source and completely #Dart. It brings all your authentication and authorization data together in the simplest way possible - a SQLite file. And it can be used in any Dart server, even outside Celest! ☁️ Self-Hosting - The new `celest build` command packages your backend in a Docker container, ready for deployment anywhere. 📈 Speed - The CLI is now 50% faster when iterating locally and we expect similar boosts when deploying to Celest Cloud, which we'll share more about later this week. More updates to come so stay tuned! Check out our updated documentation to explore all the new features, and dive into the CLI to experience them firsthand! #DartIsTheFuture 🚀 https://celest.dev/docs

I received a call from the convenor (Sambhav Kulshrestha) of our university (RNTU), who assigned me the responsibility of developing a Google Form for the forthcoming university fest (Rhythm-2024).... The stipulation entailed that upon form submission, users would receive a unique identification code.... & I provided a solution that incorporated this requirement.... btw the task itself is straightforward, but if you need a solution the source code is available on my GitHub profile repo for reference.... ------------------- 🔗 Github > selfmadechouhan > Registration-Automation.git ------------------- . . . . #UniversityFest #GoogleFormDevelopment #UniqueIDGeneration #TechnicalSolution #GitHubRepository #InnovativeApproach #EventManagement #TechSolutions #CodeDevelopment #LinkedInViral

Shared some very useful tips for GSoC 2024 in this video!! Please do check this video here👉 https://lnkd.in/eHvx5ntQ

Forget paid! 🔥 And that too for NO reason. Just because it’s free the content gets judged. At this stage I feel there are plethora of “genuine” free resources—which brings in premium quality. I feel bad for the creators who’ve come forward putting tremendous efforts in creating something impactful. Yet go unrecognized! :( Vivek Gupta, one of such exceptional teachers (ex-Google) who has made great quality lectures for FREE ⚡️ Man, checkout his teaching videos for a sec on topics like advanced DSA and CP content—you’ll realize it! The platform they built AlgoZenith includes : ~ Exhaustive compilation of >200 recent / realtime Interview Questions or tests being asked across colleges and off campus (word-to-word) 💪 ~ Weekly LIVE community learning sessions where Vivek teaches most complex DS topics in framework patterns to master them. ~ Publicly made a range of PREMIUM quality courses / workshops completely FREE—topics like binary search, C++, DP, string algorithms, segment trees, Prefix sum… If you haven’t “utilized” the platforms free resources yet, man you’re missing a lot in my opinion! 🚀 Checkout the link : https://t.ly/NFWPB P.S. Coming 2nd of march, AlgoZenith will start it’s NEW premium LIVE cohort sessions taught by none other than Vivek himself (taught for over 4+ years now) https://t.ly/qgcCS

Your contribution for tonight 🌚👇 The open-source ecosystem is filled with fantastic projects! Here are 3️⃣ really neat repos you should consider contributing to: 🤖 Co-founded by Gary Guangyu Li, AITable.ai simplifies workflow automation with a visual database. AITable.ai's main repo, apitable/apitable, is an API-oriented low-code platform for building collaborative apps. GitHub page: https://lnkd.in/gzeWX6kq 🏃♂️ Co-founded by Daniel Loreto, Jetify transforms how engineers write scalable backends on the cloud. Its main repo, jetify-com/devbox, creates isolated, reproducible development environments that run anywhere—with no Docker containers or Nix lang required. GitHub page: https://lnkd.in/ebNeKxTN 🦾 FEDML is a generative AI platform for production. It helps developers to launch complex model training, deployment, and federated learning anywhere on decentralized GPUs, multi-clouds, edge servers, and smartphones, easily, economically, and securely. FEDML was co-founded by Salman Avestimehr and Aiden Chaoyang He. GitHub page: https://lnkd.in/guSbGC6P If you want to contribute, you can head to their GitHub repo or find more of their OS stats on Quira ➡️ https://lnkd.in/eY78gDCY.