惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Cloudbric
Cloudbric
WordPress大学
WordPress大学
GbyAI
GbyAI
D
Docker
博客园_首页
有赞技术团队
有赞技术团队
Y
Y Combinator Blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
V
V2EX
博客园 - Franky
D
DataBreaches.Net
P
Proofpoint News Feed
博客园 - 司徒正美
Google DeepMind News
Google DeepMind News
H
Help Net Security
大猫的无限游戏
大猫的无限游戏
L
LINUX DO - 最新话题
K
KPMG report finds enterprise disconnect between AI and its ROI | CIO
V
Visual Studio Blog
Apple Machine Learning Research
Apple Machine Learning Research
Recent Announcements
Recent Announcements
Engineering at Meta
Engineering at Meta
The Register - Security
The Register - Security
C
Cisco Blogs
Blog — PlanetScale
Blog — PlanetScale
I
InfoQ
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
Security Archives - TechRepublic
Security Archives - TechRepublic
U
Unit 42
J
Java Code Geeks
T
Troy Hunt's Blog
T
The Exploit Database - CXSecurity.com
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
宝玉的分享
宝玉的分享
www.infosecurity-magazine.com
www.infosecurity-magazine.com
P
Palo Alto Networks Blog
H
Heimdal Security Blog
A
About on SuperTechFans
博客园 - 聂微东
Simon Willison's Weblog
Simon Willison's Weblog
S
Security @ Cisco Blogs
P
Proofpoint News Feed
B
Blog
Project Zero
Project Zero
S
SegmentFault 最新的问题
Last Week in AI
Last Week in AI
Hacker News: Ask HN
Hacker News: Ask HN
Martin Fowler
Martin Fowler
酷 壳 – CoolShell
酷 壳 – CoolShell
雷峰网
雷峰网

The New Stack | DevOps, Open Source, and Cloud Native News

Agentic development hinges on verification. For cloud-native software, that is a runtime problem. AI agents need infrastructure: Why Europe’s regional cloud strategy matters Transform your AI coding agent into a deterministic Java Spring expert WeAreDevelopers is coming to the US to give unsung developers a bigger voice Cleaner AI training data, fewer bugs: Sonar’s SonarSweep explained Observability overload is drowning engineers Google’s DiffusionGemma is 4x faster than its other Gemma models Fable 5: Guardrails and burn rate are annoying users, who say it’s still better than Opus 4.8 The Anthropic leader who built Claude Code says he ditched prompting — now he just writes loops. AWS can now mathematically prove your VMs are isolated Microsoft pulled 73 GitHub repos after malware attack — but still won’t say who’s compromised Databricks wants to kill the “email me a file” problem for AI agent skills Ramp bets forward deployed engineers can do what off-the-shelf finance AI can’t Git real: AI agents aren’t just for solo developers anymore Anthropic launches Claude Mythos/Fable 5, but you better try it soon This AI agent startup ditched Anthropic for DeepSeek — and says it’s saving millions When your data model is the bottleneck: lessons from Medium’s feature store How long before we stop reading the code? The tokenmaxxing party is over, and Revenium is mopping up How AI is solving the memory crunch it created Microsoft’s pitch to enterprises: Ditch Azure Repos for GitHub, despite its rocky reliability record Claude Code’s biggest upgrade yet ran 5 agents at once — here’s what happened Why Anthropic just doubled Claude Cowork limits at no charge For years, Apache Cassandra handed this work to your team — 6.0 takes it back “A dangerous combination”: The 2 factors that can “corrupt” AI agent workflows With Foundry, Microsoft bets the enterprise AI battle is about reliability, not capability Microsoft unlocks Visual Studio for developers left behind by its own AI AI teams now deploy 1,000 times a month. Your pipeline wasn’t built for that. Microsoft just made the agent runtime free — and kept everything around it “Whoever builds the most joyous product wins”: The agent war begins Netlify CTO Dana Lawson: Writing code is no longer the job From Jupyter Notebook to production: How to ship AI systems that actually work OpenClaw used Gavriel Cohen’s code and exposed the AI Agent accountability problem Replit shows how vibe coding is getting its own financial stack — and a path to profit Cloudflare aqui-hires VoidZero: Did a piece of the open web just stabilize, or become more brittle? Cursor cuts prices and adds enterprise spend controls amid “tokenomics” reckoning Google Gemma 4 12B nearly matches 26B benchmarks — and runs on your laptop Snowflake thinks it knows what’s really slowing developers down Autonomous agents have met their biggest challenge yet: The database. Why agentic AI makes the ops platform the most important layer in the enterprise How to dramatically improve enterprise security alert tuning to battle cyberattacks Why the need for humans won’t disappear in the age of autonomous databases How to secure Kubernetes in the age of AI workloads Asana says its new AI “chief of staff” turns your Slack chaos into trackable work Nvidia’s best model is now live Mate Security’s Asaf Wiener made every backend engineer a model router. He’s right to. The AI cost crisis finally has a watchdog — just not the companies causing it How to get operational data off the factory floor without creating an IT breach Why CPUs still matter in the age of AI agents Rayfin: Microsoft’s answer to the gap between vibe coding and enterprise production Microsoft bets the enterprise AI race will be won on data context, not model power “A successful attack could be catastrophic”: Anthropic gives more groups access to Claude Mythos How GitHub plans to win developers back Microsoft really, really, really wants developers to love Windows again With Intelligent Terminal, Microsoft is reinventing the Windows terminal Microsoft debuts “Scout” at Build, a new personal agent for work OpenAI’s Codex adds new tools — Sites, Annotations, more plugins — for knowledge workers GitHub Copilot’s usage-based billing is live: Here’s what you need to know OpenAI, Anthropic, Google, Amazon, and xAI all fail on type of attack, study finds JetBrains open-sources Mellum2 to go where Claude Code can’t Claude Code vs. Cursor vs. Codex vs. Antigravity — six months in This coding agent doesn’t want your feedback — it ships without it “Blowing things up”: The one move vendors got wrong on AI agents At Sapphire, SAP makes the case that enterprise AI is a context problem Gavriel Cohen found his own code inside OpenClaw, so he walked away AI retrieval at scale is becoming a systems problem, not a tooling problem The DIY platform trap that’s burning out engineering teams I tested Cursor’s new Jira integration and it’s 5 stars, no notes. Here’s why. Why GPT-5.4, Claude, and Gemini can’t agree on basic, real-world facts Replit’s vibe coding platform just got a Visa-backed identity layer for AI agents — and it changes how agents spend money Opus 4.8 Made Claude Smarter. Token Discipline Got Urgent. Why Linux creator Linus Torvalds gets angry hearing “99% of code is AI” Vendor neutrality isn’t magic: A hard look at the OpenTelemetry ecosystem “The AI did it” won’t save you when EU regulators come knocking The fix for soaring AI cloud bills exists — so why won’t we trust it? AI is shipping code faster than security was built to handle Why AWS scrapped OpenSearch’s architecture to chase agent workloads Claude Opus 4.8 is here: effort controls, dynamic workflows, cheaper fast mode, better honesty, less deception Percona celebrates 20th birthday with new foundation — and a goat cake Why OpenAI and Anthropic are hiring forward deployed engineer teams Claw-style AI agents are coming to the enterprise. The governance infrastructure is still catching up. The agentic identity crisis: Why your security isn’t ready for the AI revolution Debugging the undebuggable: building observability into probabilistic AI systems Snowflake commits $6B to AWS as it pushes deeper into AI Why MotherDuck refuses to fork DuckDB Researcher “gave Claude Code ‘ADHD’… and it thinks 2x better now.” Outside experts want more proof. “There is no accountability”: AI coding agents are installing packages no one owns “Tokenmaxxing is real, expensive & it’s spreading”: AI budgets are exploding With Google’s debut, the most important AI agent feature is now the most boring one Why AI agents need a Context Lake Google ranks the best AI for building Android apps, and the winner isn’t Gemini Google pushes Pro, Ultra, and free users from open-source Gemini CLI to closed-source Antigravity CLI The reason enterprise outages almost never start where ops teams think Taming the agentic influx: a blueprint for AI business observability How the AC/DC framework helps teams govern AI coding agents GitLab 19.0 trades its string section for a full DevSecOps orchestra Who’s monitoring the agents? How Jaeger hit 8.6× compression on 10 million spans with ClickHouse What ClickHouse learned from a year of coding with AI agents OpenClaw passed 300,000 GitHub stars. Then Google launched Spark.
AWS puts an AI bouncer at the merge queue
Darryl K. Taft · 2026-06-17 · via The New Stack | DevOps, Open Source, and Cloud Native News

The bottleneck in software development has moved.

 Writing code is no longer the constraint. Getting it safely to production is. AWS is targeting that shift with two new capabilities added to its AWS DevOps Agent, announced Wednesday at the AWS Summit New York.

The new release management features, now available in preview, extend the DevOps Agent beyond its existing post-deployment operations work into the delivery pipeline itself. Specifically, AWS is adding a release readiness review and an autonomous release testing capability. Both are aimed at helping review and test processes keep pace with the accelerating volume of AI-generated code.

“With so much code that is being written today by AI agents, a real bottleneck has shifted,” Neha Goswami, director of Agentic AI for Agentic DevOps at AWS, tells The New Stack. “It’s less about writing of the code, and it’s really about how to get this thing out — how do we get it out in production, and how do we get it out safely.”

Release readiness review

The first capability evaluates code changes against production requirements before they merge. The agent checks for cross-repository dependency risks that could affect other services, access control changes measured against the AWS Well-Architected Framework, and compliance with any internal standards the team has defined. Those standards are written in plain English via a Global Instructions editor. No additional tooling is required. When no standards are provided, the agent falls back to general best practices.

As part of the review, the agent runs the software in an AWS-managed isolated environment, executing lightweight user journey tests to verify the build runs and passes basic functional checks before entering the pipeline. The output is has three options: BLOCK, Proceed with Caution, or Safe to Release. Findings come out in the AWS DevOps Agent console and as comments on pull requests in GitHub or GitLab, and can also be invoked directly from the IDE via the Kiro plugin or Claude Code plugin.

The dependency-tracking dimension is what Goswami says makes the tool different from existing static analysis.

“This is going really beyond just static code analysis,” she says. “Even in the low users, we have been able to find some issues where the repo-level dependency was not being tracked by any of our existing solutions.” In one early case, a change that had already passed human review was blocked when the agent identified a cross-repository dependency conflict the reviewer had missed.

Autonomous release testing

The second capability goes further into the delivery process, generating and running change-specific test plans for web and API-based applications in customer-provisioned, production-like environments before the change merges. Rather than executing a static test suite, the agent reasons about what the change does and builds tests tailored to it. These tests cover functional correctness, behavioral regressions, and integration scenarios. Every run produces structured artifacts including metrics, logs, traces, and an execution summary.

The distinction between where each feature runs matters: the readiness review uses an AWS-managed isolated environment; the release testing runs in the customer’s own provisioned environment. That separation puts more responsibility and control on the customer side for the testing phase.

Internal scale as competitive claim

Goswami, a 22-year Amazon veteran who started in internal developer tooling, framed AWS’s competitive edge as its ability to benchmark at internal scale. AWS DevOps Agent’s release management features were tested against Amazon’s own engineering workflows before external release, with thousands of internal developers using the tooling.

“We have coded these agentic capabilities just the way we do this work,” she says, pointing to the operational excellence practices baked into how the agent decides what to test and what guardrails to apply. “A lot of what you would see is the best practices and the operational excellence that we have built into our culture, the agent acts in the same way.”

Availability

The release readiness review and autonomous release testing features are available now in preview, at no additional cost, in the US East (N. Virginia) region only. GitHub and GitLab are the supported source control platforms at launch. General availability is expected in the coming months, though AWS did not provide a specific date.

AWS DevOps Agent has been generally available for post-deployment operations, such as incident investigation, root cause analysis, and mitigation recommendations, since earlier this year. This new release marks its expansion into the pre-deployment side of the software delivery lifecycle, completing the loop from code creation to production, the company says.

YOUTUBE.COM/THENEWSTACK

Tech moves fast, don't miss an episode. Subscribe to our YouTube channel to stream all our podcasts, interviews, demos, and more.

Created with Sketch.