Experience
-
-
-
-
-
-
-
-
-
-
-
Education
-
-
-
-
Novel cost functions for Full Waveform Inversion
Supervisor: Prof. M. Warner -
-
Other similar profiles
Explore more posts
-
NVIDIA Data Center
From analyzing entire codebases to generating hours of video, some of today's AI workloads push context processing lengths past 1M+ tokens. Handling that scale requires both a new GPU and software orchestration purpose-built for the challenge. That’s where NVIDIA Dynamo and the Rubin CPX processor come together: ✅ Dynamo disaggregates inference, splitting prefill and decode phases across dedicated GPUs for maximum efficiency. ✅ #NVIDIARubin CPX is architected specifically for ultra-long context processing, unlocking higher throughput and lower latency where it matters most. The result? Performance, accuracy, and ROI—making workloads like video generation, code completion, and research-scale analysis possible at speed and scale. 🔗 Learn more: https://nvda.ws/3JYhPPC
2 Comments -
Sepp Hochreiter
gLSTM extends xLSTM to a graph neural network architecture: https://lnkd.in/da3at-Y9 "gLSTM mitigates sensitivity over-squashing and capacity over-squashing." "gLSTM achieves comfortably state of the art results on the Diameter and Eccentricity Graph Property Prediction tasks, and very strong performance on SSSP"
5 Comments -
DuckDB
We just published a deep dive on spatial joins in DuckDB by Max Gabrielsson. In the new post, Max explains how spatial joins evolved in DuckDB and how the current operator harnesses R-Trees. He presents benchmarks showing how the spatial extension in DuckDB 1.3 is now capable of performing exploratory analysis on the NYC bike rental data set. 🔗 See the link in the comments.
6 Comments -
Domyn
🔔 Product alert: Domyn Swarm is now live on Github. The new open-source toolkit is now publicly available to simplify large-scale inference for Large Language Models (LLMs) for high-performance environments. Domyn Swarm is: ✅ Built with Python and powered by vLLM ✅ Designed to run on HPC clusters like Leonardo and NVIDIA DGX Cloud ✅ Ideal for model evaluations, data generation, and rapid prototyping. 🟢 At Domyn, we tested it to parallelize our own LLM evaluation workflows, confirming its potential to save time, reduce complexity, and significantly boost iteration speed. Special thanks to CINECA for their continued support and commitment to advancing open research! ➡️ Read our news story: https://lnkd.in/dATHM6qV ➡️ Get Domyn Swarm on GitHub: https://lnkd.in/dw-XaG5P
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Nikhil Shah in United Kingdom
-
Nikhil Shah
-
Nikhil Shah
Energy Integration Advocate | Projects | R&D | Energy Consultant
-
Nikhil Shah
-
Nikhil Shah
-
Nikhil Shah
68 others named Nikhil Shah in United Kingdom are on LinkedIn
See others named Nikhil Shah