Click any tag below to further narrow down your results
Links
This is Anthropic's hands-on course teaching you how to write better prompts for Claude, with 9 chapters covering prompt structure, common failure modes, and practical techniques. You work through exercises and test examples in a live playground to see how prompt changes affect Claude's responses.
- The course teaches 80/20 techniques to fix common prompt failures and covers Claude's specific strengths and weaknesses
- It's structured as 9 chapters plus an appendix, designed to be worked through in order with practice exercises and an answer key
- The tutorial uses Claude 3 Haiku (the smallest and cheapest model) but notes that Sonnet and Opus are available as more capable alternatives
Pinterest replaced its multi-stage clustering pipeline with a single 4-billion-parameter LLM that reads user activity logs and outputs ranked journey names—goals like "Modern Farmhouse Kitchen" rather than generic interests. They distilled this from expensive frontier models into a deployable student model using synthetic training data.
- The old clustering system fragmented related activities (cabinets, countertops, backsplash treated as separate journeys) and fell back to generic labels like "Art" or "Shopping" when clusters were messy. Collapsing the pipeline into one generative model solved this by reading the full activity sequence at once.
- Prompt engineering became the ranking and safety policy—all the logic previously encoded in separate model weights and thresholds now lives in explicit instructions about engagement strength, recency, coherence, and specificity. Changes to prompts are treated as launches requiring review discipline.
- The 4B student model achieves frontier-model quality at scale through distillation: frontier models generate synthetic training examples from real activity histories, then the smaller model learns to replicate those outputs at inference cost Pinterest can actually afford for hundreds of millions of users.
@mikenevermiss on X
A short X post from @mikenevermiss claims one prompt is enough to get Claude to architect a full "Jev decision layer," and urges readers to save it. The post itself is a teaser that quotes an linked article, so the substance lives in that quoted piece rather than in the post text.
- The core claim is that a single prompt pasted into Claude can produce a complete decision layer design.
- The author frames it as a reusable recipe worth saving, not a detailed walkthrough.
- The post links to a quoted X article, which holds whatever method or prompt is actually being described.
Using LLMs directly as classifiers is frustrating because they can't be calibrated, don't reliably use all available data, and are hard to improve systematically. Instead, treat LLM outputs as features in a logistic regression or other ML model, which solves these problems while keeping the LLM's signal.
- Direct LLM classification fails on three critical fronts: you can't calibrate confidence scores or adjust precision-recall tradeoffs, the model may ignore structured data or context you provide, and you have no visibility into what the LLM actually used to make its decision.
- Wrapping an LLM verdict in logistic regression automatically fixes calibration (empirically matching true probabilities), lets you add other features, and gives interpretability about how much the LLM contributes to the final decision.
- On an irony detection task, logistic regression on the raw LLM verdict improved Brier score from 0.259 to 0.175 just by calibrating; adding more LLM-extracted features (token probabilities, multiple reasoning passes) provides a clear path to further improvement that prompt-tweaking doesn't offer.
@wulfie_bain_ on X
An OpenAI Startups applied AI lead argues that prompts in even the best AI companies grow contradictory over time because teams only add lines and never remove them. He proposes treating prompts like code and product decisions, using a modular, MECE structure (Background, Behaviour, Output).
- Prompts evolve accretively, so contradictions and ambiguity pile up. Conditional prompts spread across files make the whole thing hard for anyone to review end to end.
- He reports results from prompt rewrites: one startup cut agent speed-ups of about 50%, another saw 40% higher 7-day retention, and another cut costs by 30%.
- Smarter models don't fix ambiguity. Someone still has to specify what the product should do, and evals force that clarity.
@AnatoliKopadze on X
This is a promotional X post pointing to a 38-minute video where OpenAI CEO Sam Altman talks about using LLMs, paired with a guide on building a system that prompts itself. The post itself offers no substance beyond the video pitch and the claim that the talk beats 99% of users.
- Altman's core claim, as quoted, is that people will no longer need to write prompts.
- The post promotes a follow-up guide on building a self-prompting system, but the guide's contents are not included here.
- The "better than 99% of people" framing is unsupported marketing language, not a measured result.
Most LLM failures come from poorly written system prompts, not model selection. This guide covers the specific structures (XML tags for Claude, markdown for GPT), token budgeting strategies, and five patterns that consistently work—like making roles absurdly specific, showing exact output schemas instead of describing them, and explicitly telling models when to admit they don't know something. Testing matters too: run your prompts through happy paths, edge cases, and adversarial inputs multiple times before deploying.
- A 400-token system prompt outperformed a bloated 3,000-token version because the model got confused parsing everything at once
- Showing exact JSON schemas instead of describing them jumped format compliance from 70% to 95%+
- Claude performs 15-20% better with XML tags while GPT prefers markdown with JSON schemas and Gemini prefers markdown with numbered rules
- One ambiguous instruction (like "include all relevant context" next to a JSON format requirement) can cause a week of debugging
The author argues that treating prompts as disposable is wasteful—instead, build modular "skills" (organized folders with instructions and scripts) that you improve over time and chain together. He breaks down 17 tactics from Anthropic engineers into 6 layers, showing how to structure workflows so improvements compound across tasks rather than starting from scratch each time.
- The author rebuilt a 600-line SEO prompt into six modular skill files (research, audit, metadata, internal links, GSC monitoring, title scoring), and improving one now improves every workflow that uses it.
- The core shift is from one-off prompting to building reusable "skills"—folders with a SKILL.md file, scripts, and references—that Claude runs identically every time and that get refined after each failure.
- Anthropic engineers' 17 tactics get grouped into 6 layers, with the top priority being: stop writing custom prompts for repetitive tasks and instead build slash-command-invoked skills.
- The lasting value comes from the infrastructure built on top of Claude, not from any single prompt, which evaporates once the session ends.
@eng_khairallah1 on X
A post claims that a file Andrej Karpathy reportedly uses, called LOOPS.md, transformed the author's Claude setup, making responses feel tailored to how they think. The author's pitch is to build a system that drives the model rather than prompting it turn by turn.
- The author says a friend on Karpathy's team at Anthropic shared the LOOPS.md file he actually uses, five weeks after he joined.
- After adding the file to their setup, the author reports Claude's first response was "completely different" and matched their working style.
- The core takeaway is that users should build a persistent system around the model instead of talking to it directly.
@Raytar on X
A post from X user @raytar claims a friend who is an Anthropic engineer shared an internal video on how the core team prompts Claude, and says the author realized after watching it that they had been using Claude wrong for two years. The post itself contains no concrete prompting techniques, so the value depends entirely on the linked video.
- The author says a friend at Anthropic, who they claim earns $1.2 million a year, sent them an internal video of the core team's prompting playbook.
- The author describes the video as something "never supposed to get out" and says nothing better on prompting exists.
- The post names no specific technique, example, or finding, so it is mainly a pointer to the video rather than a standalone source of guidance.
@asmah2107 on X
The post is a short endorsement from @asmah2107 saying a two-week trial of a custom system prompt improved their AI results, and it quotes Marc Andreessen (@pmarca) posting that prompt. The prompt tells the model to act as a world-class expert, verify facts, skip disclaimers and moral commentary, avoid flattery, hold its position under pushback, and give long, detailed answers.
- The prompt is a single block of instructions covering persona, verification, tone, and format, meant to be prefixed to every query.
- Key directives: no hallucination ("if you don't know, say so"), no disclaimers or ethics lectures unless asked, lead with the strongest counterargument, and don't cave to pushback without new evidence.
- The poster's only evidence is personal: "works better always" after two weeks of use. No benchmarks or comparisons are given.
Spotify built Vedder, an AI data assistant that uses domain-specific “clusters” curated by expert teams to provide reliable SQL-generated insights. Each cluster combines relevant datasets, vetted question–SQL pairs, and docs, with automated health scoring and feedback loops to keep context accurate and scalable.
- Auto-generating question-SQL examples from query logs mostly failed—only 12.5% passed expert review, proving curated human examples beat scraped ones
- Vedder's reliability comes from domain-expert-curated "clusters" (tables + vetted SQL examples + docs), not a bigger LLM
- Health scoring tracks schema drift and broken examples so experts know which clusters need fixing as data evolves
- Adopted by 2,100+ employees for 13,000+ conversations since August 2025, showing real traction beyond a pilot
@anujcodes_21 on X
Anthropic has published a free 27-minute workshop on writing prompts for Claude, taught by the team that built the model. The post's author argues it covers more practical ground in its first few minutes than paid courses costing $300 do in full.
- Anthropic's workshop is 27 minutes long and free to watch, with no registration or paywall.
- The instructors are the people who built Claude, not third-party course creators.
- The author claims paid $300 prompting courses cover less in their first 8 minutes than this workshop does overall.
This article examines how AGENTS.md files impact AI coding agents, showing that well-structured agent docs can boost code quality by up to 15% while poorly designed ones can hurt performance. It outlines seven patterns that work—like progressive disclosure, step-by-step workflows, decision tables, real-code examples, and pairing “don’ts” with “dos”—and warns against overexploration from excessive context or warnings.
- AGENTS.md quality swings are huge: good ones boost output as much as upgrading from Haiku to Opus, bad ones perform worse than having no doc at all.
- Keep it short (100-150 lines) with details pushed to reference docs—bloated files cause "overexploration" where agents burn tens of thousands of tokens on irrelevant context and produce worse results (one case: completeness dropped 25% after the agent read 80K extraneous tokens for a two-line config change).
- Concrete structure wins: numbered step-by-step workflows cut missing files from 40% to 10% (+25% correctness), decision tables boosted best_practices by 25%, and real 3-10 line code snippets raised code_reuse by 20%.
- Always pair "don'ts" with concrete "dos," and scope docs to modular subsystems (~100 files) rather than writing sprawling repo-wide guides.
This article breaks down the key sections and best practices for your CLAUDE.md file, the project context guide that Claude Code reads before writing or editing code. It covers what to include—from project overview and tech stack to safety rules and commands—so the AI stays aligned with your repo’s structure and standards.
- CLAUDE.md is read by Claude Code before every code write/edit, so it functions as persistent project context rather than a one-time config file
- Specificity beats generality: naming exact versions/libraries ("Next.js App Router + TypeScript + Tailwind + shadcn/ui + Supabase") works far better than vague stack descriptions
- Concrete coding rules like "use named exports except in route files" or "avoid any; prefer explicit interfaces" produce better-fitting code than generic guidance like "clean code"
- The file should act as a living reference to guide suggestions, not a rigid enforcement mechanism that blocks workflow
The author argues that modular “Skills”—reusable markdown workflows loaded on demand—outperform standalone AI agents by cutting token bloat and maintenance overhead. A live GEO audit system built with Skills shows how you can turn domain expertise into scalable, service-ready products without managing dozens of agents.
- Claude's "Skills" load modular markdown playbooks on demand instead of baking everything into prompts, citing 53 tokens for passive reference vs. embedding a full prompt every time
- A live GEO audit system built entirely on Skills scrapes visibility across ChatGPT/Gemini, flags gaps like missing Wikipedia entries, and auto-generates client-ready reports without spinning up separate agents
- The whole GEO pipeline is public and forkable, letting anyone productize it without building custom infrastructure
- Documenting expertise once in a markdown file and iterating on it lets teams ship service-ready AI products in days rather than maintaining fleets of bespoke agents
This article examines how Claude Code constructs its system prompts by analyzing recently leaked source code. It highlights the dynamic nature of these prompts, which are built from various components based on conditional statements. The author also outlines additional elements that contribute to context, like tool definitions and conversation history.
- Claude Code's system prompt isn't static text—it's dynamically assembled at runtime from conditional components, revealed by a recent source code leak.
- The prompt mixes always-included elements (like core instructions) with conditional ones (like tool-specific guidance) that only appear based on context.
- Beyond the system prompt itself, the full context sent to the model includes tool definitions, conversation history, attachments, and skills—all engineered together to shape responses.
The article explores the parallels between the film "Lord of War" and the current AI compute market, focusing on Jensen Huang of Nvidia as the central figure in the AI arms race. It details an experiment where multiple AI image generation models were tested to recreate a parody poster, "Lord of Tokens," using advanced prompts that challenge the models' capabilities. The results highlight varying levels of success in achieving the desired artistic and technical details.
- The "Lord of Tokens" poster parody frames Jensen Huang as the central arms dealer of the AI compute boom, mirroring Nicolas Cage's character in "Lord of War"
- Multiple leading AI image generation models were torture-tested with a complex, design-agency-grade prompt to recreate this parody poster
- The models showed significantly varying success rates in nailing the specific artistic composition and technical details demanded by the prompt