The Workstation Blog
All Posts
Hands-on guides to AI infrastructure, MLOps, models, and shipping to production.
Vector Search Acceleration
Abstract Generative AI applications require similarity search over embeddings stored alongside operational data. This article compares in-PostgreSQL vector…
From the Hall at GTC Berlin: Huang Says AI Is Now Productive
Workstation engineers were at STATION-Berlin for the 21 October 2026 keynote. Huang's turn was from factories to agents that spend tokens.
Balinder WaliaRead OpenShell, Announced at GTC Berlin: Policy Outside the Agent
Huang's security film rejected 'please don't' as a control. OpenShell enforces a policy on syscalls, files and network, with Sentry on BlueField-4.
Balinder WaliaRead 
The Berlin Slide That Splits Documents from Databases
Nemotron Parse, Embed, cuVS and Rerank on one row. Auto-Ontology, cuDF and Kumo on the other. Huang's supplier-delay example uses both.
Balinder WaliaRead 
DSX, the CUDA Wheel, and Europe's AI Stack at GTC Berlin
Huang's factory half: DSX from a digital twin, Vera Rubin pods, and a slide that stacks applications, models, factories, NVIDIA compute and energy.
Balinder WaliaRead NVIDIA Minerva for AI Factories: Turning Alert Storms into Recovered GPU Hours
One degrading optic can raise thousands of alerts and stall a thousand GPUs. What NVIDIA Minerva does, what it is worth in GPU hours, goodput, MTBI, MTTR and tokens, which of its three consumption models fits, worked examples for a Vera Rubin NVL72 neocloud, a university Blackwell cluster and an enterprise with Grafana and ServiceNow, and how it relates to UFM.
Balinder WaliaRead 
NVIDIA UFM for AI Clusters: Why InfiniBand Fabric Visibility Decides What Your GPUs Are Worth
One flapping InfiniBand link can stall a 256-GPU job. What NVIDIA Unified Fabric Manager does, when UFM Telemetry, Enterprise or Cyber-AI fits, worked examples for a 4-node DGX B300 cluster, a 32-node SuperPOD-style cluster and a university Slurm service, and what fabric visibility is worth in GPU hours.
Balinder WaliaRead 
Build AI Agents with NebulaDB: Semantic Search, RAG and a Visa AI Agent
How we built semantic search, RAG and a Thai Visa AI Agent on NebulaDB, with real queries, query plans and retrieval scores measured on our 8.3M-document production server, and the guards that keep private data and invented numbers out of the answers.
Balinder WaliaRead 
NVIDIA Blackwell Ultra B300 Has Landed in the UK: Configure Yours Today
NVIDIA Blackwell Ultra B300 systems have arrived with our UK distribution partners. Configure and price your eight-GPU B300 build in minutes with our AI Sales Engineer, save with Workstation Trade, and order today to be first in line as UK supply arrives.
Balinder WaliaRead 
Private AI for Law Firms: Choosing an Offline-First Workstation Your Clients Can Trust
How a law firm chooses a private-AI workstation that runs with no internet connection: who uses it, which model, how much GPU memory, how isolated it must be and who sets it up. With real prices from a real quote, where it was over-specified, and the Starter, Workforce and Enterprise builds we would actually ship.
Balinder WaliaRead