Lance Martin

@RLanceMartin

MTS @anthropicai

San Francisco, CA

rlancemartin.github.io

Replies Reposts Media Articles

Lance Martin
@RLanceMartin
Sep 2
i recently added this command to the claude-api skill. run it in Claude Code to fix common prompting "anti-patterns" that can hobble frontier models: /claude-api prompt-audit patterns include: 1/Verification rituals. Instructions like "double-check your work” or "verify
Peter Yang
@petergyang
Sep 2
If you're trying out Fable 5.1 I highly recommend running: /claude-api prompt-audit on your skills. It finds a bunch of redundancies and rules to remove for the latest models. Running it for all my skills now.
60
Lance Martin
@RLanceMartin
Sep 1
some tips for Claude Fable 5.1 Try Fable 5.1 at low effort | Fable 5.1 at low effort is often competitive with Opus and Sonnet on $ / task while scoring higher. on CursorBench 3.2.0, Fable 5.1 at low effort is at parity w/ Fable 5 at high effort at a third of the cost. Prompt
71
Lance Martin
@RLanceMartin
Jul 26
lol this talk is from AIE 2024 (before I was at Anthropic). im glad graphs are cool (again)!
Codez
@0xCodez
Jul 25
Anthropic engineer just released a 2-hour workshop on "Graph Engineering" for agentic systems: “80% of our engineers are using self-improving loops. Now everyone is building agentic graphs.” • 00:00 - Introduction to RAG & Graphs • 06:39 - Core of "Graph Engineering" (state,
00:00
14
Lance Martin
@RLanceMartin
Jul 24
Opus 5 is a great model for async / long-horizon work. Check out my talk from @aiDotEngineer on a few useful patterns : 1) split the brain + hands, 2) self-correction loops, 3) memory with dreaming, 4) org-level agent harnesses.
Claude for Long-Horizon Tasks — Lance Martin, Anthropic
From youtube.com
6
Lance Martin
@RLanceMartin
Jul 21
this is a nice talk from Vercel Ship 2026. Harrison (a core contributor on Claude Managed Agents) covers some of the benefits of splitting the "brain" (agent harness) from the "hands" (@vercel sandboxes inside your VPC):
4