Claude Blog 采集 (2026-07-22)¶
共采集 5 篇文章
📋 文章索引¶
- How Datadog built a “universal machine tool” for Claude Code - Jul 21, 2026 (评分: 9.5)
- How Anthropic secures its AI-native software development lifecycle - Jul 21, 2026 (评分: 9.0)
- Meet the winners of the Built with Opus 4.7 Claude Code hackathon - Jun 15, 2026 (评分: 9.0)
- Meet the winners of our Built with Opus 4.6 Claude Code hackathon - Apr 20, 2026 (评分: 9.0)
- Meet the winners of our Claude Opus 4.8 Build Day hackathon - Jun 17, 2026 (评分: 9.0)
How Datadog built a “universal machine tool” for Claude Code¶
来源: Claude Blog 发布日期: Jul 21, 2026 采集时间: 2026-07-22 价值评分: 9.5/10 正文字数: ~6082 字符
摘要¶
Datadog has an agent write specifications for a deterministic kernel to write application code
正文内容¶
Datadog has an agent write specifications for a deterministic kernel to write application code.
Agents, mechanization, and industrialization All of Datadog engineers use AI coding tools for production code, and Claude Code drives at least two-thirds of that. With Claude Code, they generate personalized flows in their software development lifecycle in four distinct categories: Targeted changes: dozens of gnarly bug fixes, performance optimizations, and bridges to existing services. Large refactors: refactoring a custom protobuf parser in three days as well as rewriting a metrics control from FoundationDB to Postgres in under three months. Replacing large parts: new sharding algorithms and autoscaling redesigns. Building entire systems : replacing MongoDB with Postgres, BYOC control planes, and ingestion pipelines from scratch. As work flowed across this map, however, they saw it became more complex to generate on one axis and more ambiguous to verify on the other. The flow problem For engineers, flow used to mean a direct relationship between intent and code. You understood the problem, wrote the code, tested it, reviewed it, shipped it, operated it, repeated. With agents, the abstraction is changing rapidly. “You're no longer writing the code; you're shaping the work. You're deciding what the agent should see. What tools it should have, what success means, how failure should be detected…It's like everyone's promoted three levels up into the management chain, which they didn’t sign up for because they're engineers,” says Sesh Nalla, VP of engineering, Datadog. With approaches like Claude Managed Agents , Datadog’s sessions run longer, sometimes for days. Each agent invents its own tools, its own glue code, and its own conventions. The agents become significantly more useful, but need humans to bridge the gap between agent execution and tools designed for humans. Machine tools are the jigs, fixtures, gauges, and mills you see in manufacturing. They produce precise, repeatable parts that you assemble into larger, more complex machines like engines, aircraft, nuclear reactors, and lunar landing modules. They were the breakthrough of industrialization as parts became composable, inspectable, and replaceable. Temper is what Sesh describes as Datadog’s attempt at a universal machine tool for agentic systems. In other words, the smallest kernel required for agents to build what they need in a safe and precise manner. “This is the point where I felt we needed something more structural,” says Sesh. “If agents are going to build and operate large parts of our systems, of our databases, which are mission critical, they need the equivalent of this machine tool concept. Temper is that machine tool for Datadog.”
The road to Temper Mechanization means agents are doing more of the work now. And industrialization means work becomes repeatable, verifiable, controllable, and scalable. At Datadog, this didn’t happen all at once: the path to Temper led through three other projects, Courier, BitsEvolve, and Helix. Each one exposed the bottleneck for the next, and enabled them to grow their ambition. In 2024, they introduced Courier , a distributed queuing system. It took them one year to build completely by hand and from scratch. “The difficulty was not building the parts; it was making the interactions between them observable, testable, and verifiable,” says Sesh. “So we were rigorous with formal modeling and simulation… identified the parts where mistakes would be expensive or hard to reverse, and raised the rigor [there].” In September 2025, they built BitsEvolve , a closed-loop evolutionary optimization harness. A council of models generates code variants. A cascade of benchmarks , tests, and production observability decides what survives. “This was the first glimpse for me that parts of software could be cultivated like living organisms — grown through variation with feedback, and adaptation,” says Sesh. The catch: evolution is only as good as the environment it adapts within, and BitsEvolve’s bottleneck was this feedback loop. Then they built Helix , a Kafka-comparable streaming service. Claude Code did most of the construction with one human steering it. “To our disbelief, in a few days we had a fully functional Kafka comparable system,” says Sesh. “[It was quick to build] and we started shadowing it and we saw opportunities where it could be 2x to 5x cheaper.” Getting it to production, though, took a lot more mileage: the operational hardening only earned over time and by more than one person and this is still in the process of rolling out. “The bottleneck moved again where agents could build large parts of the system…but then humans still have to coordinate to ship the work to production through tools and mechanisms built for humans,” says Sesh. Datadog needed a way for agents to build their own tools in a verified, policy-driven runtime environment. That runtime was Temper . Temper Agents can produce code faster than any team can review by hand, but they can make mistakes. For Sesh, that gap between what an agent generates and what passes verification is where the failure modes accumulate. However, simply wrapping an agent around a traditional codebase treats this as a throughput problem without closing the verification gap itself. Temper reverses this equation: instead of producing application code, agents produce specifications. The kernel reads each specification, verifies it through four layers of analysis, and deploys the running system the specification describes. Because the specification is both the artifact that gets proved and the artifact that gets executed, there is no drift between what was verified and what is running.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
How Anthropic secures its AI-native software development lifecycle¶
来源: Claude Blog 发布日期: Jul 21, 2026 采集时间: 2026-07-22 价值评分: 9.0/10 正文字数: ~3467 字符
摘要¶
Anthropic Deputy CISO, Jason Clinton, details how the Security Engineering team secures a SDLC that has AI authoring 80% of merged code.
正文内容¶
Anthropic Deputy CISO, Jason Clinton, details how the Security Engineering team secures a SDLC that has AI authoring 80% of merged code.
At Anthropic, the amount of code and velocity of deployment have scaled exponentially. Our software engineers on average ship 8x as much code per quarter as they did from 2021 to 2025. Our reviews, monitoring, and other security processes needed to scale alongside this increased pace. Otherwise it becomes a formula for bottlenecks ( Amdahl’s Law ). Our software development processes have changed drastically as well. Claude has evolved from coding assistant to primary creator and reviewer. Claude authors about 80% of the code merged into our codebase today. More than half of all code is being merged by our internal version of Claude Tag while human engineers focus on directing, setting intent, and owning final approval. This means our security team must defend a rapidly expanding surface area and harden a lifecycle with non-deterministic, constantly evolving agents at its heart. In this article, I cover strategies to secure the software development lifecycle (SDLC). (This is intended to be combined with the Zero Trust for Agents framework we recently published; everything in this article uses security design ideas from that framework in the implementation). The threats we're designing against are specific: a compromised or prompt-injected agent introducing a malicious change; supply-chain and dependency poisoning that an agent ingests as trusted input; and the more familiar classes of application vulnerability now arriving at higher volume. Every control that follows maps to at least one of those. There are several overarching strategies we’ve deployed to accomplish this without significantly throttling dev velocity including: Shifting security left and fully integrating with the code development stage; Using hard access and identity boundaries to contain the blast radius; Combining automated deterministic and agentic reviews before and after production; and Inserting humans in the loop at the highest leveraged points. In this article, we’ll cover the security processes we have implemented at specific stages of the software development lifecycle as well as the core principles behind them. These principles are more enduring as security teams must reexamine, and often reinvent, their processes as model capabilities evolve. The evolving software development lifecycle
Code Security professionals within an AI-native engineering organization have a new lever: they can directly shape how code is created, helping to prevent vulnerabilities at the source. Previously, teams observed recurring vulnerabilities and created secure coding guidelines to address them, but those guidelines were difficult to enforce and rarely standardized. At Anthropic, those guidelines are encoded in CLAUDE.md files and references to org-wide skills so the code follows these best practices the minute it's generated. This is done as part of a closed loop. Once an agent discovers a bug class, the relevant file is updated to prevent it recurring in future code.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Meet the winners of the Built with Opus 4.7 Claude Code hackathon¶
来源: Claude Blog 发布日期: Jun 15, 2026 采集时间: 2026-07-22 价值评分: 9.0/10 正文字数: ~1311 字符
摘要¶
We chatted with the winners of our Built with Opus 4.7 hackathon about their projects, tackling medical training, electronics repair, computer science education, interactive play, home repair, and factory maintenance.
正文内容¶
From medical training and electronics repair to coding education and factory maintenance, see the projects built by the winners of our latest virtual hackathon.
Last week, we hosted Claude Build Day , our latest hackathon where builders got together in San Francisco to put their ideas to work using Claude Opus 4.8. While we wait to see what they built, we chatted with the winners of our Built with Opus 4.7 hackathon about their projects. They tackled medical training, electronics repair, computer science education, interactive play, home repair, and factory maintenance. Congratulations to the winners and to everyone who participated! We hope their ideas will inspire you. First place: Medkit , Bedirhan Keskin Bedirhan Keskin, an Istanbul-based physician-turned-software engineer, used Claude Managed Agents to build Medkit: a learning tool for medical residents or junior doctors, simulating real-life patient encounters in a gamified medical clinic.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Meet the winners of our Built with Opus 4.6 Claude Code hackathon¶
来源: Claude Blog 发布日期: Apr 20, 2026 采集时间: 2026-07-22 价值评分: 9.0/10 正文字数: ~1425 字符
摘要¶
From a cardiologist to an electronic musician, get to the know the winners of our Built with Opus 4.6 hackathon.
正文内容¶
From a cardiologist to an electronic musician, get to the know the winners of our Built with Opus 4.6 hackathon.
Last week, we announced the Built with Opus 4.7 virtual Claude Code hackathon , inviting the community to build with our latest Opus model. As with our previous hackathon featuring Opus 4.6, we’re partnering with Cerebral Valley to select 500 participants and give each $500 API credits and one week to build with Claude Code. Judges from the Claude team will then pick six winners and award them from a total prize pool of $100,000 in Claude API credits for project development. The winners of our Opus 4.6 hackathon, our first in this series, included a personal injury lawyer, a cardiologist, a roads and infrastructure specialist, an electronic musician, and one professional software engineer. They tackled projects to improve housing, healthcare, infrastructure, music, and education. And four out of five winners were not professional developers. We hope their projects will inspire you to build something meaningful. First place: CrossBeam , Mike Brown
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Meet the winners of our Claude Opus 4.8 Build Day hackathon¶
来源: Claude Blog 发布日期: Jun 17, 2026 采集时间: 2026-07-22 价值评分: 9.0/10 正文字数: ~1070 字符
摘要¶
From reconstructing Tang Dynasty architecture to polling a synthetic San Francisco, see what the winners of our latest hackathon built with Claude Opus 4.8 in a day.
正文内容¶
From reconstructing Tang Dynasty architecture to polling a synthetic San Francisco, see what the winners of our latest hackathon built with Claude Opus 4.8 in a day.
On June 13, we brought more than 300 founders and builders to San Francisco for a 12-hour hackathon with Claude Opus 4.8. More than 1,500 people had applied; 310 took part, many traveling from around the world, each with $500 in credits and one day to turn an idea into a working demo. We caught up with the three winning teams about what they built and how they used Claude to do it. Congratulations to the winners and everyone who took part. We hope their projects give you a few ideas of your own. First place: Tekton , Holly Tang and Austin Burgess
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集