Co-founder of @SpareCores, @RxStudioInc; @ceu lecturer;
#rstats#python#data#emacs#linux#qtile maniac;
Ex @system1co, @CARD, @rapporters;
Father of 👦👧👧👶
The final program of #erum2018 is now online at https://t.co/4WB1vzQa2d
The recent updates include all the #rstats keynote, invited, regular and lightning talks, also the poster and Shiny demo abstracts, detailed schedule, social program etc.
Registration closes in ~2 weeks!
I'm thrilled to be speaking at @pydatalondon🇬🇧 in two weeks: choosing a cloud server shouldn't be a guessing game; it's an actual data problem. @SpareCores presents an open cloud server dataset on hardware, performance & cost efficiency right from your #pydata or #rstats console.
Are you overprovisioning cloud resources for your ML jobs? Chances are, yes 💸
In his talk at PDAmstedam2025, @daroczig shows how the resource-tracker package can help.
Full talk: https://t.co/UDDlCKM3fG
#PDAmsterdam2026 Super Early Bird tickets: https://t.co/Dr8z0ntDJK
Year-end reflections usually mean deep thoughts, heavy quotes, and bold ROI claims .. but since 2025 was all about AI, I wrapped it up by vibe-coding a simple, high-value, scalable app: a binary countdown to New Year’s Eve: https://t.co/oLdqv4kb2e 🤖🥂
Gergely Daróczi on reviving Budapest Users of R Network (BURN) + using R with Strava data for daily insights—his personal diabetes story. https://t.co/2W2sIRjy8B
#rstats#opensource#Strava#rprogramming
We benchmarked 2k+ cloud servers for LLM inference speed (prompt processing and text generation) using models ranging from 135M to 70B parameters!🤖
Read tech details, use our open-source tools, learn from the results at https://t.co/6t1d6XZE9s
#LLM#benchmarks#cloud
🎤 Excited to speak at #KCDBudapest next Thu!
I'll share how we spun up 2,000+ cloud servers at @SpareCores to benchmark LLM inference speed from 135M to 70B params, 16–16k tokens. 🚀
Public data & visualizations, and I'd love to chat with interested parties. Come say hi! 😊👋
Ever wondered how much CPU, memory, GPU, disk, or network traffic your @MetaflowOSS steps use and how to optimize resource allocation? The open-source resource-tracker Python package does that (and more) with seamless #metaflow integration! Read more at https://t.co/dxt0XVH6Qw 🚀
Filter & rank 2000+ cloud servers by 500+ benchmarks, including memory bandwidth, OpenSSL speed, PassMark, GeekBench, Redis, static web serving and more! LLM inference speed workloads also coming soon. Check it out! 👉 https://t.co/rDlaytc7A0
We are excited to welcome @UpCloud and their 67 public cloud server offerings to our platform - tracking pricing, hardware specs, and various benchmark workloads in a standardized way. Discover more at https://t.co/uYStYOD0bY
On my way to @OSXP_Paris to represent the @SpareCores team -- please say hi at booth C20 (with @ow2 and @NgiSearch) or join my talk at 12.40 tomorrow if you are around (photo attached to make it easier to find me)! I'm super excited about the next 3 days 😊
I'm hopping on a night train in a few mins ... and after a quick 12 hours, I'll arrive at Bolzano! Can't want to connect with fellow open-source enthusiasts at @SFScon over the next few days 😊
PS I brought a few @SpareCores hex stickers to share!!
🗣️ Gergely Daroczi (@daroczig) from @SpareCores will present "Optimizing Cloud Compute Resources with Spare Cores" at #SFSCON24.
📌 https://t.co/dNxOkrBrWf
🎫 Join & register for free: https://t.co/Q46pOgwWHC
#SFSCON#NOITechpark