4. Sandwich long documents with your task. When pasting 50+ pages, place the task BEFORE the doc AND restate it AFTER. [task] → [document] → [restate task] 1M context handles the length, but task proximity to the doc boundary locks the model's attention on what you actually
AI
-
Optimizing LLM response quality using XML tags as attention anchors
By
–
3. Switch from markdown to XML tags. Opus 4.7 responds to XML structure the way other models respond to markdown headers. Wrap your sections in , , , . The model treats them as attention anchors. Same content, dramatically better
-
Optimizing LLM Prompts by Removing ‘Think Step by Step’ Scaffolding
By
–
1. Delete "think step by step" from your prompts. That phrase was scaffolding that compensated for 4.6's reasoning gap. Opus 4.7 has that depth built in at high effort. Keeping it now wastes tokens AND can degrade output. Fix: raise the effort level instead.
-
Improving AI Prompt Precision for Better Model Performance
By
–
2. Stop writing prompts that assume the model "gets it." 4.6 prompt: "help me with my pitch deck"
(4.6 inferred you wanted structure + hooks + content tips) 4.7 prompt: "review the 10 slides below. flag weak hooks, missing stats, and confusing transitions. output as a -
Adjusting Prompting Strategies for Opus 4.7 Behavior Changes
By
–
The shift is simple but devastating: Opus 4.6 inferred your intent and filled in the gaps.
Opus 4.7 does EXACTLY what you ask. Nothing more. Translation: the same prompt now produces narrower, terser, sometimes broken results. Here's how to fix it -
Prompt Engineering Adjustments for Claude 3.7 Opus
By
–
RIP your old Claude prompts Opus 4.7 dropped 3 weeks ago and it follows instructions LITERALLY now. It scores 87.6% on SWE-bench (vs 80.8% on 4.6) but every prompt tuned for 4.6 is silently failing. 7 fixes that stopped my outputs from getting worse:
-
Agentic coding vs software engineering
By
–
This means that agentic coding isn't exactly a replacement for software engineering. It is a fundamentally different way of producing software, with different best practices and different use cases. Just like ML.
-
Agentic coding as machine learning
By
–
Agentic coding is a form of machine learning. Generated code is best treated as a blackbox artifact whose behavior and generalization should be managed via empirical evaluation, like with any ML model.
-
Scaling Senior Engineer Productivity with Claude Code and Multi-Agent Workflows
By
–
Así escala de verdad un Senior Engineer con Claude Code.
— Nico (@nicos_ai) 9 mai 2026
La diferencia está en mover tu tiempo hacia lo que más importa:
→ mejores prompts, más planificación, más review, menos tecleo.
El workflow:
usa un plugin que divide cada tarea entre 5 agentes:
– uno hace brainstorming
-… pic.twitter.com/PkMxnKfmEJHere's how a Senior Engineer truly scales with Claude Code. The difference lies in shifting your time toward what matters most:
→ better prompts, more planning, more review, less typing. The workflow:
use a plugin that divides each task among 5 agents:
– one does brainstorming -
Gary Marcus on AGI and scaling limits
By
–
So many people misremember (or never read) what I said in in 2022 in “Deep learning is hitting a wall”, which was neither about revenue or AI’s potential upper limits. Rather, it was an argument that the pure of scaling LLMs would not get us to AGI, and that we would need to