Stories tagged “Software Engineering.”
15 stories
Grok Bot vs. Hermes: Where each draws the security boundary
The New Stack compares Grok Bot and Hermes, two AI agents with different security boundaries. While Grok Bot is designed to be more secure, Hermes is more focused on ease of use.
Kubernetes at the edge has hit a wall. Fleet management is the way through.
Kubernetes at the edge has hit a wall, but fleet management is the way through. The New Stack explores the challenges and solutions in this article.
AMD's AI systems have achieved a 30 percent overall productivity boost through AI, surpassing their initial target of a 25 percent boost. They are rethinking the structure of the software development lifecycle (SDLC) and believe the biggest AI revolution in software engin…
Ask HN: What are you working on? (August 2026)
Users share their current projects and what they're working on, including a carpentry simulator with an agent MCP.
Block, a fintech company, raised its full-year outlook after reporting a strong second quarter driven by growth in Cash App and Square. The company also expanded its use of AI across software engineering.
Harry Goldstein, Editor in Chief of IEEE Spectrum, reflects on the retirement of Robert N. Charette, a renowned risk management and software engineering expert. Charette, known for his insights into software failures and the automation paradox, served as a long-time contr…
Researchers propose CORVUS, a novel trajectory architecture for LLM coding agents that decouples file-read actions from their observations, reducing redundant file copies and stale snapshots.
A benchmarking study on Opus 5's performance on SlopCodeBench, a new long-horizon coding benchmark. The study found that Opus 5 got a 24% strict pass rate, but failed to reach the final checkpoint with no defects.
Cursor, Ramp, and Meta are all building model routers — but two have major model ambitions themselves
Cursor, Ramp, and Meta are building model routers, but two of them have major model ambitions themselves. This development is significant for the software engineering community, as it indicates a shift towards more complex and ambitious projects.
Moonshot launched Kimi K3. Then demand shut down subscriptions in 48 hours.
The New Stack launched a moonshot project called Kimi K3, but demand for the project was so low that they shut down subscriptions in just 48 hours.
The bottleneck for AI agents isn’t the model anymore. It’s the context layer.
The bottleneck for AI agents isn't the model anymore. It's the context layer. The New Stack explores the shift in focus from model development to context layer infrastructure.
OpenAI hits 8 million Codex users — what developers need to know
OpenAI has reached 8 million users for its Codex platform, a significant milestone in the development of AI-powered tools for software engineering. This growth highlights the increasing adoption of AI in the industry and the potential for Codex to revolutionize the way de…
Open-source AI is just '4 months behind' closed frontier models — and 10x cheaper
The New Stack reports that open-source AI models are only four months behind their closed frontier counterparts, with the added benefit of being 10 times cheaper. This development has significant implications for the software engineering industry, particularly for compani…
APIs aren’t dead. Here’s where MCP fits alongside them.
The article discusses the role of MCP in incident management alongside APIs, emphasizing that APIs are not dead. It highlights the importance of MCP in managing incidents and its benefits.
GitHub Copilot's agentic harness is evaluated across models and tasks, showing efficient performance and token efficiency. The harness powers various GitHub Copilot experiences and is designed to be fast, token-efficient, and predictable.