JoJo's Inspirations

JoJo's Inspirations #30 Favikon Top Creator 🔥 with 253K+ followers. Agentic Thought Leader. Keynote Speaker and Board Member. Host of The Generative AI Revolution Podcast.
(2)

Publisher of Women in Tech Chronicles Magazine. Multiple Patents and Multiple OTA Awards.

08/07/2026

Which RAG stage is your AI system really at?

RAG has evolved beyond simple retrieval.

1️⃣ Naive RAG
• Basic retrieval and generation
• Good for FAQs
• Limited accuracy

2️⃣ Advanced RAG
• Hybrid search
• Re-ranking
• Query rewriting
• Better precision

3️⃣ Agentic RAG
• AI agents choose the best retrieval path
• Multi-step reasoning
• Tool selection

4️⃣ GraphRAG
• Combines vector search with knowledge graphs
• Understands relationships, not just similarity

Each stage solves what the previous one couldn't.

Choose the stage that fits your problem—not the hype.

💬 Which stage is your RAG system using today?

08/05/2026

AI agents didn't replace APIs.

They added two new protocols on top.

The foundation is still the same:

🔹 REST – Public APIs
🔹 GraphQL – Flexible client queries
🔹 gRPC – High-speed internal services

The new agent layer introduces:

🤖 MCP – Lets AI agents connect to tools using one open standard.
🔄 A2A – Lets AI agents discover, communicate, and hand off work to each other.

The simplest way to think about it:

• REST, GraphQL & gRPC = Services talking to services.
• MCP = Agents talking to tools.
• A2A = Agents talking to agents.

Choose the protocol based on who's communicating, not what's newest.

→ Save this as your AI protocol cheat sheet.

→ I break down AI systems like this every week in The AI Edge: https://lnkd.in/gfnJ6jmJ

💬 Which is your team exploring first: MCP or A2A?

Views my own.

08/03/2026

Enterprise AI agents no longer have to start from scratch.

IBM watsonx Orchestrate now includes 6 pre-built Domain Agents ready to connect to your business systems.

Here's what's available:

👥 HR – Time off, hiring, job transfers
💼 Sales – Prospecting, follow-ups, CRM prep
📦 Procurement – Vendors, contracts, policies
📊 Finance – Planning, forecasting, scenario analysis
🚚 Supply Chain – Inventory, orders, disruptions
🎧 Customer Service – Phone, chat, and messaging

The real advantage?

All of them run under one control plane, with centralized routing, governance, audit, and monitoring.

As you scale AI, you don't lose visibility or control.

Build vs. buy just became a much more interesting conversation.

→ Save this for your next AI planning session.

→ I share practical enterprise AI insights every week in The AI Edge: https://lnkd.in/gfnJ6jmJ

💬 Which business function would you automate first?

Views my own.

Hot pink, no explanation needed. 💗Showed up for the grad. Stayed for the whole celebration. 🎓Some evenings just deserve ...
08/02/2026

Hot pink, no explanation needed. 💗
Showed up for the grad. Stayed for the whole celebration. 🎓
Some evenings just deserve the good outfit.

07/31/2026

There’s a difference between using AI and building it.

IBM’s AI Engineering Professional Certificate helps bridge that gap.

Inside the 13-course Coursera program, you’ll learn:

🧠 Machine learning & deep learning
🤖 Neural networks & transformers
📚 RAG & AI agents with LangChain
💻 Python, PyTorch & Keras through hands-on labs

This is best for intermediate learners with basic Python knowledge.

You can also audit the courses for free if you don’t need the certificate.

🔗 https://lnkd.in/eQnsNHJy

→ Save this for your learning list.

→ I share practical AI resources every week in The AI Edge: https://lnkd.in/gfnJ6jmJ

💬 Which skill do you want to learn next: RAG, AI Agents, or Transformers?

Views my own.

Today’s agenda: enjoy the sunshine and appreciate the artist who somehow turned golden hour into a hair color. Thank you...
07/31/2026

Today’s agenda: enjoy the sunshine and appreciate the artist who somehow turned golden hour into a hair color. Thank you, Elona Taki at Taki Salon & Spa . ☀️

07/29/2026

Everyone is building AI agents.

Few are measuring whether they actually work.

A chatbot gives one answer.

An AI agent makes dozens of decisions behind the scenes.

That’s why evaluation matters.

📊 Measure more than accuracy:

→ Accuracy
→ Latency
→ Cost
→ Hallucinations
→ Tool-call quality
→ Task completion

Accuracy alone won’t tell you if an agent is production-ready.

Reliable AI comes from measuring every step, not just the final answer.

→ Save this as your AI agent evaluation checklist.

💬 What’s the hardest part of evaluating AI agents today?

→ I share practical AI tools and resources every week in The AI Edge: https://lnkd.in/gfnJ6jmJ

Views my own.

07/27/2026

Most people learn AI agents by reading about them.

The fastest learners run the code.

IBM has open-sourced a collection of free, runnable notebooks for Granite models that work in Google Colab. No local setup required.

Inside you’ll find practical examples for:

🤖 AI agents (ReAct & Agentic RAG)
📚 RAG with LangChain, LlamaIndex & Docling
🔧 Tool & function calling
🔗 Model Context Protocol (MCP)
📊 Model evaluation
🛡️ AI safety & guardrails

These aren’t tutorials.

They’re production-ready patterns you can try yourself.

One hour of hands-on practice beats weeks of reading.

🔗 https://lnkd.in/gMK5KfgQ

→ Save this for later.

→ I share practical AI resources every week in The AI Edge: https://lnkd.in/gfnJ6jmJ

💬 Which would you try first: RAG, Agents, or Evaluation?

Views my own.

07/24/2026

Most people don't know where to start learning AI.

IBM created a free AI learning pathway that meets you where you are.

Whether you're a beginner or building advanced skills, the path grows with you.

📚 Start with:
• AI fundamentals
• Prompt engineering
• AI ethics

🚀 Then move into:
• Data analytics
• Fraud detection
• watsonx

⚙️ Advance to:
• Machine learning
• NLP
• AI deployment

You don't need to learn everything at once.

The best way to learn AI is one step at a time.

🔗 Explore the pathway: https://lnkd.in/gyyxK2wv

→ Save this for your learning list.

→ I share practical AI resources every week in The AI Edge: https://lnkd.in/gfnJ6jmJ

💬 Where are you today: Beginner, Intermediate, or Advanced?

Views my own.

07/22/2026

One idea from Chapter 3 changed how I think about AI:

You can't think your way into an AI mindset.
You have to act your way into it.

Most people start by collecting information.

Another video.

Another course.

Another newsletter.

They keep learning about AI instead of learning with AI.

The shift happens when you stop waiting to understand everything...
..and start experimenting.

The people I've seen get the most from AI weren't the smartest.

They were the most willing to try.

To fail.

To learn.

To keep going.

You don't build confidence before you begin.

You build confidence by beginning.

💬 What's one thing you finally stopped overthinking and just started doing?

Address

777 Woodward Avenue, Suite 800
Detroit, MI
48226

Website

Alerts

Be the first to know and let us send you an email when JoJo's Inspirations posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Contact The Business

Send a message to JoJo's Inspirations:

Shortcuts

Share

Category