Alexandre Contador

Alexandre ContadorAug 18, 2026

A tool that strips AI watermarks from your files just hit 4,700 GitHub stars in days. Guillaume Meyer released a tool called watermarks-remover under the MIT license. In a few days, it passed 4,700 stars on GitHub. What it does: strips invisible AI markers from files. Not just plain text. It works on PNG, JPEG, SVG, PDF, DOCX, HTML, and Markdown. Meyer is upfront about the tradeoff. Rewrite too aggressively, and you weaken the watermark signal. But you also flatten tone, accuracy, and the quality of the original writing. He's also clear on the limits. No tool can guarantee it beats an official manufacturer's detector. This isn't sold as a perfect solution. Comment bel... jk, here you have it: https://lnkd.in/e9cBM3-V This is an easy one!

Alexandre ContadorAug 19, 2026

Most people don't have a Claude problem. They have a prompting problem. Same model. Same subscription. Wildly different output. Over the past year I've built a prompt library covering every function I work across as a fractional CGO — sales, marketing, product, ops, finance, HR, C-suite. 200+ prompts, 45 skills, all of it from client work rather than theory. The carousel covers four of the eight principles. The highest-leverage one is the simplest: most people write requests when they should be writing specs. "Write a sales email." vs. "I'm a B2B SaaS founder selling project management software to mid-market engineering teams (50–200 people). Write a cold email to a VP of Engineering whose company just raised Series B. Goal: book a 20-minute demo. Tone: peer-to-peer, not salesy. Max 150 words." Same ask. Only one gets you something you'd actually send. The library is free — no course, no upsell. Comment PROMPTS, and I'll send the link.

Alexandre ContadorAug 24, 2026

ChatGPT doesn't rank websites. It remembers them. That distinction matters more than most marketers realize. Google rewards freshness and links. AI answers reward authority and trust - built over time, across sources. Here's what actually gets your brand surfaced in ChatGPT, Perplexity, and every AI that comes after them: 1. Get cited, not just indexed LLMs are trained on what the web talks about. If trusted sources mention your brand - G2 reviews, industry blogs, Substack posts, Reddit threads - you exist in the model's memory. If you only exist on your own site, you're invisible. 2. Own your entity What does your brand stand for in one sentence? If that sentence isn't consistent across your website, your LinkedIn, your reviews, and third-party mentions - the model gets confused and doesn't surface you. Clarity beats cleverness every time. 3. Be the answer to a question ChatGPT returns information in response to questions. If your content doesn't answer specific questions your buyers are asking - not keywords, actual questions - you won't be referenced. Write for humans who are talking, not humans who are typing. 4. Show up in the places LLMs trust Quora. G2. Capterra. Trustpilot. Wikipedia. Reddit (yes, don't outbid yourself there due to updates) These sources are heavily weighted because they're harder to game. 5. Get mentioned in comparison content "Best tools for X." "Alternatives to Y." "X vs Z." These listicles are catnip for AI search. If you're not in them, you're not in the conversation. The good news: none of this changes with model updates. Authority, trust, and clear positioning have always mattered. AI just made them non-negotiable. What are you doing to show up in AI answers? Ps.: Try Rankahead to get this fully on auto-pilot #GEO #AISearch #GrowthMarketing