Claude Code: Best practices for agentic coding(anthropic.com)

549 points by sqs 1 year ago | 232 comments

Anthropic engineering blog post sharing tips and best practices for using Claude Code, a command-line tool for agentic coding, across codebases and languages.

Building Effective "Agents"(anthropic.com)

572 points by jascha_eng 1 year ago | 100 comments

Anthropic blog post sharing practical advice and simple composable patterns for developers building effective LLM agents, based on work with customers.

The ways we contain Claude across products(anthropic.com)

212 points by jbredeche 4 months ago | 92 comments

A postmortem of three recent issues(anthropic.com)

361 points by moatmoat 1 year ago | 112 comments

Anthropic engineering blog post giving a technical postmortem of three bugs that intermittently degraded Claude's responses, and what is changing.

Effective context engineering for AI agents(anthropic.com)

128 points by epenson 1 year ago | 24 comments

We tasked Opus 4.6 using agent teams to build a C Compiler(anthropic.com)

690 points by modeless 8 months ago | 664 comments

Anthropic engineering blog post describing how teams of parallel Claude Opus 4.6 agents built a C compiler with minimal human involvement.

Agent design is still hard(lucumr.pocoo.org)

383 points by the_mitsuhiko 11 months ago | 219 comments

Armin Ronacher's blog post on lessons learned building AI agents, covering SDK choice, caching, reinforcement, failure isolation, shared state, and model choice.

AI demands more engineering discipline. Not less(charitydotwtf.substack.com)

321 points by BerislavLopac 4 months ago | 157 comments

Charity Majors' Substack essay arguing that AI-assisted development requires more engineering discipline, comparing it to the shift from pet servers to immutable infrastructure.

Exploiting the most prominent AI agent benchmarks(rdi.berkeley.edu)

561 points by Anon84 6 months ago | 136 comments

UC Berkeley RDI blog post describing how researchers exploited major AI agent benchmarks like SWE-bench and Terminal-Bench to score highly without solving tasks, and proposed fixes.

Don't build AI products the way everyone else is doing it(builder.io)

469 points by tortilla 3 years ago | 219 comments

Builder.io blog post by Steve Sewell arguing against building AI products as thin LLM wrappers and describing an alternative approach to building them.

The new skill in AI is not prompting, it's context engineering(philschmid.de)

832 points by robotswantdata 1 year ago | 467 comments

Blog post by Philipp Schmid arguing that context engineering, supplying LLMs with the right information and tools, matters more than prompting for AI agents.

Building better AI tools(hazelweakly.me)

320 points by eternalreturn 1 year ago | 165 comments

Blog post by Hazel Weakly arguing the industry builds AI tools backwards, drawing on research about how humans learn and transfer knowledge.