Security Hardening for Claude Code in Large Repos
Sandboxing, least privilege, secret handling, and prompt-injection defense for running Claude Code safely against real production codebases.
7 min read
Agentic AI & LLMs
Agentic AI & LLMs archive page 66 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
Cutting Claude agent costs: caching, batching, cheaper runs
Keep Claude agent runs cheap and fast with prompt caching, the Batches API, model routing, and ruthless context trimming. A founder's cost playbook.
7 min read
Cutting Claude Code Token Cost in Large Codebases
Prompt caching, batching, and context discipline that keep Claude Code runs cheap and fast on large codebases without sacrificing agentic work quality.
7 min read
Debugging Claude agents: loops, bad tool calls, hallucinated args (Founders Playbook AI Native Startup)
Diagnose the three big Claude agent failure modes — loops, wrong tool calls, and hallucinated args — with reproducible traces and boundary validation.
8 min read
Debugging Claude Code in Large Codebases: Failure Modes
Why Claude Code loops, picks the wrong tool, or hallucinates arguments in large codebases — plus the concrete fixes that get agentic runs back on track.
8 min read
Context Design for Claude Code in Large Codebases
What to put in Claude Code's context and what to leave out in large codebases: memory tiers, retrieve over paste, attention budgeting, and the dilution trap.
7 min read
Prompt and context design for Claude agents that work (Founders Playbook AI Native Startup)
What to put in a Claude agent's context, what to leave out, and why — a founder's guide to context design that makes agents reliable, not flaky.
7 min read
Wiring MCP servers into a Claude agent the right way
Production patterns for connecting MCP servers to Claude: auth, schemas, error handling, and idempotency — the wiring that keeps agents reliable.
7 min read
Wiring MCP Servers Into Claude Code the Right Way (Claude Code Large Codebases)
Connect MCP servers to Claude Code safely in large repos: scoped auth, constraining schemas, recoverable errors, and idempotency that survives retries.
7 min read