Skip to content
360SOFTY
HomeInsights

Insights

Ideas for building
better software.

Engineering perspectives on architecture, product development, AI, and the everyday decisions behind useful software.

From the engineering desk

Browse practical articles or follow the latest technology updates.

Subscribe via RSS →

Curated links from external sources — not 360Softy original articles.

ExternalAI
NVIDIA Technical Blog

Dense vs. MoE Models: Active Parameters, Throughput, and When to Choose Each

How can a 30B-parameter model activate only 3B parameters per token, and still use the capacity of the larger model? Nemotron 3.5 Lightning illustrates the... How can a 30B-parameter model activate only 3B parameters per token, and still use the capacity of the larger model? Nemotron 3.5 Lightning illustrates the answer: It uses a Mixture-of-Experts (MoE) architecture that selects only a subset of its parameters for each token. There are two dominant model architectures: Dense model and MoE. How

NVIDIA Technical BlogRead original
ExternalSoftware Engineering
DZone

KV Cache vs Prompt Cache: What's the Difference, and How Are They Related?

This article was originally published on my blog. For the latest version and future updates, please visit the original post: https://jaketao.com/language/en/kv-cache-vs-prompt-cache/. Every time a large language model generates a token, it draws on the content that came before it. If it had to compute everything from scratch at every step, responses would be much slower. When building an agent, the same set of system prompts, tool definitions, and conversation history is used over and over again

DZoneRead original
ExternalSoftware Engineering
DZone

Agentic AI Threat Intelligence Essentials

Agentic AI can help threat intelligence teams move faster across collection, enrichment, correlation, and drafting while keeping critical judgments in human hands. This Refcard covers the essentials of building agent-supported intelligence workflows, including how to preserve provenance, evaluate source reliability and confidence, produce analyst-ready outputs, and apply review and operating guardrails. You’ll also learn how to measure whether agent-supported intelligence is timely, relevant, an

ai
DZoneRead original
ExternalAI
NVIDIA Technical Blog

How NVIDIA Groq 3 LPX Deterministic Execution Drives Power-Efficient High-Interactivity Inference on NVIDIA Vera Rubin

Power is a defining constraint for AI factories. As AI workloads demand a full compute platform to serve them, each component of that platform must maximize... Power is a defining constraint for AI factories. As AI workloads demand a full compute platform to serve them, each component of that platform must maximize output within the factory’s limited power budget. This makes performance per watt—rather than raw, unnormalized throughput—the ultimate measure of an AI platform’s value. The NVIDIA V

NVIDIA Technical BlogRead original
ExternalAI
NVIDIA Technical Blog

How NVIDIA NVLink 6 Delivers Multi-Layer Resiliency for AI Factories

For operators of large-scale AI factories, maximizing continuous output is essential for productivity. In massive-scale AI training, every GPU in the cluster... For operators of large-scale AI factories, maximizing continuous output is essential for productivity. In massive-scale AI training, every GPU in the cluster must synchronize gradients across thousands of collective operations per second. Similarly, during inference, unplanned downtime directly reduces the total volume of requests served

NVIDIA Technical BlogRead original
ExternalSoftware Engineering
DZone

Getting Started With Agentic AI for SecOps

Agentic AI can help SecOps teams investigate alerts by gathering context, correlating signals, and surfacing evidence-backed recommendations, without handing over control of response actions. This Refcard walks through how to build a bounded alert triage agent, from defining the use case and mapping inputs to configuring read-only tools, guardrails, and human escalation. You'll also learn how to structure analyst-ready outputs and test the workflow against sample alerts before piloting it in pro

ai
DZoneRead original
ExternalSoftware Engineering
GitHub Changelog

SHA-1 in HTTPS on GitHub sunset

Previously we announced we would be disabling SHA-1 in HTTPS for GitHub on September 15th, 2026. Following the planned schedule, GitHub has disabled SHA-1 in HTTPS for github.com and partner… The post SHA-1 in HTTPS on GitHub sunset appeared first on The GitHub Blog.

GitHub ChangelogRead original

Let’s start with a conversation

Tell us what you’re working on.

An idea, a challenge, or a system that needs to work better. We’ll help you understand the next step.

Discuss your project

Prefer email? [email protected]