Bogdan Sulima

Web Developer

[Graphsignal @GraphsignalAI]

"you only have like 15% of the chip being utilized." - @AnjneyMidha

Getting more out of the GPU is exactly what we're working on at @GraphsignalAI, particularly with auto-flags: https://t.co/p0EPPsTUZI

(Our founder @dmitrimelikyan in the background at the @tbpn podcast recording for additional validation 😄)

[Graphsignal @GraphsignalAI]

Profiling CUDA in production inference is a different problem than profiling a dev run.

What actually works:

https://t.co/dMifgncFPt

[Graphsignal @GraphsignalAI]

autodebug: an autonomous loop that deploys an inference service, benchmarks it, reads profiling telemetry, and redeploys with a better config. Then repeats.

Uses @GraphsignalAI for inference profiling, @dstackai for GPU provisioning, Claude Code as the agent.

https://t.co/HHdHa6TcaM https://t.co/IsIBt9hbel

[dstack @dstackai]

Now @GraphsignalAI integrates with dstack — add @sgl_project profiling, tracing, and GPU metrics to your inference services.

pip install 'graphsignal[cu12]' + wrap with graphsignal-run. That's it.

https://t.co/TEFptiG1ak

[Mario Nawfal @MarioNawfal]

This gave me chills...

[Graphsignal @GraphsignalAI]

Exciting news today! @GraphsignalAI can now automatically trace and monitor 🦙 LlamaIndex @gpt_index applications in production and development.

Details here: https://t.co/c6qbckDovy

[Graphsignal @GraphsignalAI]

Monitor and trace LangChain performance with Graphsignal!
https://t.co/QuO53u1cPM
#openai #gpt3

[Dmitri Melikyan @dmitrimelikyan]

Did an experiment with prompting #ChatGPT to generate #Prolog program to solve a simple logical problem (instead of hallucinating an answer) and then running it in Prolog interpreter. Kind of works, except too much creativity😀. Works better with low temperature in Playground.

[Bogdan Sulima @bonya]

Check vars_prompt #Ansible for interactive ansible scripts http://t.co/247G7vtxIO

[Bogdan Sulima @bonya]

Hacking #Ansible: ANSIBLE_KEEP_REMOTE_FILES=1 http://t.co/LtV2mo6e0G

[Bogdan Sulima @bonya]

Fig vs Vagrant is like Heroku vs AWS #Vagrant #Docker #DevOps https://t.co/WT28wG9KSR

[Bogdan Sulima @bonya]

ggplot for python http://t.co/Yp3bFojFzi via @yhathq

[Bogdan Sulima @bonya]

Statistics vs Data Science vs BI http://t.co/I2U9dBc5np via @prismatic

[Bogdan Sulima @bonya]

Performance Anomaly Detection with Nodetime http://t.co/2bp5GxQePF #NodeJS #Performance #Monitoring

[Bogdan Sulima @bonya]

Facebook for iOS gets ability to update cover photos on the fly, group messaging refinements - The N http://t.co/UTwUnlAnX8

[Bogdan Sulima @bonya]

Congratulations @nodetime! in february - 5 billions data points collected http://t.co/AndST9sKF7 #nodejs #performace #monitoring

[MonsterDE @MonsterDE]

Was Freiberufler beachten müssen: Welche Gesetze sich 2013 ändern http://t.co/zMRdKGWh (CW) #Karriere RT @jochenfischer #job #geld

[Bogdan Sulima @bonya]

Levels of Data Analysis: http://t.co/E71jT0Ja via @swgoof

[Bogdan Sulima @bonya]

http://t.co/LBuzRpBs the data set of 53.5 billion HTTP requests

[Bogdan Sulima @bonya]

google http://t.co/67IMoQ9D search engine