Dogukan Uraz TunainITNEXTOpenAI’s Code Execution Runtime & Replicating Sandboxing InfrastructureLooking into OpenAI’s code execution runtime for Assistants API and replicating its sandboxing infrastructure on Google Kubernetes Engine.Jul 2Jul 2
Dogukan Uraz TunaUtilizing Airflow for Planning, Scheduling, Executing and Scaling AI Agentic WorkloadsMy summarized thought about this convergence: Despite having such a setup, a toy agent setup, as a conclusion, I believe the Airflow engine…May 6May 6
Dogukan Uraz TunaLinen Modules for Simple Long Convolutions Sequence Modeling with Log-Linear Time ComplexityIntroducing the LongConv-Jax. Implementation of the “Simple Long Convolutions Sequence Modeling” paper in Jax library.Mar 20Mar 20
Dogukan Uraz TunaReleasing Tri-RMSNorm: Yielding Speed-up for Training Stabilization + [Let’s Review] Root Mean…In this blog post, releasing accelerated Tri-RMSNorm compute kernel v0.1.0 and going into the details of LayerNorm and RMSNorm processes.Mar 14Mar 14
Dogukan Uraz Tuna[Let’s Review] Gradient Ascent + Jax ImplementationIntroduction to gradient ascent algorithm and its use in Jax/FLAX models through GradientAscent-Jax packaged, implemented by simudt.Mar 11Mar 11
Dogukan Uraz Tuna[Let’s Review] Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient LMsArchitectural review of the paper “Griffin” by Google Deepmind and implemetation in Jax.Mar 71Mar 71
Dogukan Uraz TunainQTurkeyKTHack 2020KTHack2020 kuantum teknolojileriyle ilgilenen meraklıları, uzmanları ve akademisyenleri buluşturuyor.Sep 22, 2020Sep 22, 2020
![Releasing Tri-RMSNorm: Yielding Speed-up for Training Stabilization + [Let’s Review] Root Mean…](https://miro.medium.com/v2/resize:fill:160:106/1*DgGtucazuDENURsYEuWivQ.png)
![Releasing Tri-RMSNorm: Yielding Speed-up for Training Stabilization + [Let’s Review] Root Mean…](https://miro.medium.com/v2/resize:fill:320:214/1*DgGtucazuDENURsYEuWivQ.png)
![[Let’s Review] Gradient Ascent + Jax Implementation](https://miro.medium.com/v2/resize:fill:160:106/1*3AEXbPy-tJg6nh6N9r2CWQ.png)
![[Let’s Review] Gradient Ascent + Jax Implementation](https://miro.medium.com/v2/resize:fill:320:214/1*3AEXbPy-tJg6nh6N9r2CWQ.png)
![[Let’s Review] Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient LMs](https://miro.medium.com/v2/resize:fill:160:106/1*P_7Fw7xxrS9WBncw23k8IQ.png)
![[Let’s Review] Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient LMs](https://miro.medium.com/v2/resize:fill:320:214/1*P_7Fw7xxrS9WBncw23k8IQ.png)
