Huge quality of life upgrade for devs: We've added automatic prompt caching to the API which means you no longer have to set cache points in your requests!
PROMPT ENGINEERING
-
Repeating Prompts Boosts Non-Reasoning Model Performance
By
–
now trending on alphaXiv when using non-reasoning model, simply repeating your prompt improves performance. basically 1-10% absolute improvement, depending on prompt ordering or the model. And this applies to every model, it wouldn't create extra latency or generated
-
Three Ways to Optimize LangSmith Agent Memory and Performance
By
–
LangSmith Agent Builder uses memory to improve with feedback. Three practical ways to get the most out of memory: → Tell your agent to remember what works
→ Use skills to give it specialized context when needed
→ Edit its instructions directly when that's faster Full -
Fine-tune AI Models Without GPU Using Claude Code and Unsloth
By
–
Great application of skills: post-train models directly from Claude Code or Codex. No code, no GPU required. We fine-tuned LFM2.5-1.2B-Instruct with just a prompt 👀 Ben Burtenshaw (@ben_burtenshaw) You can finetune AI models with Unsloth + Hugging Face, and right now it’s free! We're giving away GPU credits on HF Jobs + Unsloth. Train a 1.2B model like @liquidai LFM2.5-1.2B with one command. Ship it to your phone, laptop, or API. No infra. No setup. Just prompt your coding agent and go. — https://nitter.net/ben_burtenshaw/status/2024552060558229858#m
→ View original post on X — @maximelabonne, 2026-02-19 18:47 UTC
-
Avoiding One-Shot Solutions: Multi-Process Testing Approach
By
–
My advice is to not try to get it to one-shot it. Actually I have a pet theory that very stern instruction primes a junior developer role (because that's who gets micromanaged). Just invoke a second process, it's better for the tests not to have the same context anyway
-
Claude Max subscription limitations outside code environment
By
–
This is why you can't use your max sub outside of claude code
-
Claude Sonnet 4.6 for Google Ads Research and Creation
By
–
For many tasks, Claude Code is still my go-to. But here is one use case: I just used it as a task speed test, for something neither I nor my team have done before. I asked Sonnet 4.6 while on a blank page – “hey go and research Google ad best practices and then create a paid
-
Coding Assistants’ Limited View of Correctness and Long-Term Code Quality
By
–
Coding assistants don't get much reward signal for long-term considerations, so they view correctness narrowly. It's easy to wave off style debates and the idea of "anti-patterns", but there's an important idea at the heart of it.
-
Framework for effective AI prompting
By
–
The difference between these prompts and random AI use: → You give it a role with real stakes
→ You force it to be honest, not polite
→ You ask for specific output, not general advice
→ You treat it like a senior expert, not a search bar That's the whole framework. -
AI-generated personal MBA curriculum with prompts
By
–
R.I.P Harvard MBA. I built a personal MBA using 12 prompts across Claude and Gemini. It teaches business strategy, growth tactics, and pricing psychology better than any $200K degree. Here's every prompt you can copy & paste: