Scaling Claude Code from one team to many cleanly
Patterns from a Built-with-Opus hackathon for scaling agentic coding with Claude across an organization — a thin shared spine, champions, and inherited guardrails.
7 min read
Agentic AI & LLMs
Agentic AI & LLMs archive page 268 of 631
Agentic AI & LLMs
Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Latest articles
When to use Claude Code agents — and when not to
Honest trade-offs from a Built-with-Opus hackathon: when agentic coding with Claude pays off, when it doesn't, and the simpler alternatives to choose instead.
7 min read
Governance and guardrails before scaling Claude Code
The governance, trust, and safety guardrails leadership needs before scaling agentic coding with Claude — permission boundaries, data controls, and audit logging.
7 min read
Team adoption of Claude Code: habits that actually stick
Change-management lessons from a Built-with-Opus hackathon: the habits, norms, and review rituals that make Claude Code adoption durable across a team.
7 min read
The real ROI of a Claude Code hackathon: cost model
A concrete cost model for Claude Code and Opus 4.8 ROI — where time and money savings actually come from, and the loops that quietly erase them.
7 min read
Migrating a Workflow to Claude Code Agents Without Breaking It
A hackathon-tested playbook for moving an existing workflow onto Claude Code agents — shadow runs, canary cutover, rollback, and parity evals.
8 min read
Evals for Claude Agents: Measuring Quality & Gating Releases
A hackathon playbook for testing Opus 4.8 agents — eval sets, rubric scoring, LLM judges, and an automated eval loop that gates every release.
8 min read
Security Hardening for Claude Code Agents: A Field Guide
Harden Opus 4.8 agents — sandboxing, least privilege, secrets handling, and prompt-injection defense from a Built-with-Opus hackathon.
8 min read
Cutting Claude Code Token Cost: Caching & Batching Wins
Keep Opus 4.8 agent runs cheap and fast — prompt caching, batching tool calls, and context pruning from a Built-with-Opus hackathon.
8 min read