Karim Chaanine
@BrandGrowthOS
CMO at Papa Johns (ME) | AI obsessive | Built Erika & Ava (15+ tools) | Co-Founder http://FitEcho.ai | Testing new LLMs & sharing my journey of creating Agents
Most AI automation content is theoretical. Mine comes from deploying AI at enterprise scale. As CMO at Papa Johns, I've built two AI agents in production: 🤖 Erika: Handles marketing finance via Outlook. 8 months running. 🤖 Ava: Personal assistant connected to 15+ tools. When…
Friday reflection: The AI landscape moves so fast that strategic planning feels impossible. But that's exactly why it's essential. This week brought new model announcements, API updates, and shifting competitive dynamics. The temptation is to chase every new capability, but…
The difference between AI automation that works and AI automation that transforms your business comes down to workflow design. I've implemented dozens of AI workflows across marketing, operations, and customer service. The most successful ones share three characteristics: clear…
Can't wait to test it out!
The wait is over! Meet Step 3 — the groundbreaking multimodal LLM from StepFun! 🚀 MoE architecture (321B total params, 38B active) 💡 Rivals OpenAI o3, Gemini 2.5 Pro, and Claude Opus 4 in performance 🖥️ Optimized for China’s domestic AI chips StepFun just announced: Step 3…
Reports suggest GPT-5 could launch as early as August 2025. If you're running enterprise AI operations, now is the time to prepare your infrastructure. GPT-5 will likely bring significant improvements in reasoning, multimodal capabilities, and context handling. But the real…
OpenAI DevDay 2025 is happening this October, and the enterprise implications are massive. Last year's DevDay introduced GPTs and the Assistants API - tools that fundamentally changed how businesses deploy AI. This year's focus on "building the future of AI together" suggests…
Ohhh boy... @cursor_ai , @windsurf_ai , @lovable_dev , @boltdotnew , ready for some more action? 😂
Today we’re releasing GitHub Spark — a new tool in Copilot that turns your ideas into full-stack apps, entirely in natural language.
just migrated 3 sales analysts from claude sonnet 4 to kimi k2 initially thought the model was broken - sales data outputs were inconsistent and causing errors then realized I forgot to set temperature to 0.1 for deterministic responses once adjusted, everything worked…
small but huge QoL improvement when you're 50+ nodes deep into a complex workflow, descriptive names become critical for debugging and team handoffs this saves hours of manual renaming across my 47+ production workflows. Well played @n8n_io
Automatic node naming is here ⚡ Tired of manually renaming nodes? Now you don’t have to. n8n updates node names automatically based on the resource & operation, so you always know what each one does at a glance. More context. Faster building. Fewer clicks. 🔗 Learn more:…
The new ChatGPT Agent export feature is a game-changer for enterprise teams. Being able to export deep research reports as formatted PDFs with tables, images, and linked citations transforms how we handle strategic analysis. I've been testing this with our market research…
Well Kimi K2 hype is short-lived. Gotta love competition..
Qwen just released Qwen3-Coder, a 480B-parameter MoE with 35B active parameters which supports the context length of 256K tokens natively and 1M tokens with extrapolation methods, showing Sonnet-like performance on benchmarks. Apache 2.0, available on Hugging Face Hub. Test it…
Andrew Ng's new course on Retrieval Augmented Generation (RAG) couldn't come at a better time. Enterprise leaders are realizing that AI's real power isn't in generating content—it's in making sense of existing knowledge. At Papa Johns, I've seen how RAG systems can transform…
n8n just dropped their simplified social media posting node, and combined with their new AI agent-to-agent features, this is a game-changer for marketing automation. Here's what I'm building: AI agents that can analyze campaign performance, generate content variations, and…
