In partnership with
Date: 2-Feb-2026
Hey {{first_name | AI enthusiast}},
You have likely faced this - you go through X/Twitter, bookmarking juicy posts that you hope to come back and read later. But you never do. Because life happens and some other task gets on your agenda.
This weekend, I had a few hours of that and decided to package it all up into the latest edition of One more thing in Ai newsletter. That was a forcing function for me to read and understand and share it with you.
Plus: Nisha Pillai, our guest writer is back with a free and incredibly useful tutorial on how to build your own AI Chief of Staff. You can’t miss this one if you want to make yourself 100x more productive!
Table of Contents
Hope you enjoy this.
PS: If you want to unleash the power of AI agents to grow your business, setup time speak to me, here»
Guest Column- AI & Error by Nisha Pillai
Build Your Own AI Chief of Staff (A Tutorial & Starter Pack)
Everyone's building an "AI Chief of Staff" these days. You’ve heard about mine in earlier editions of this newsletter. In the past few months, I've seen dozens of demos of other AI Chiefs of Staff. Many are to-do lists with AI polish. Some are calendar assistants. Others are meeting prep tools. A few are full-blown agent systems that require API keys and integrations. None of them are wrong, exactly. But none of them were what I had built.
What I wanted when I started building Leo was simpler: an AI that actually knows me. My goals, not generic productivity advice. My constraints, not best practices from someone else's life. My patterns—including the ones I'd rather not admit to. Something that could ask me the right questions because it understood the context I'd already given it. As you’ve heard about, that’s not an easy problem to solve.
I've been writing about Leo here for months—the failures, the rebuilds, the moments when it finally clicked. Leo started as an experiment and has become something I use every single day. This week, I’m giving you a starter pack and a tutorial to get you started on building you own.
What This Is
I’ve published this as a free, public, course that walks you through building a Personal Chief of Staff as an LLM Project (I use Claude). No coding. No security issues. No permissions beyond what Claude already has. Just you, a Claude Pro subscription, and some setup time spread across the first week.This isn’t tied to Claude, feel free to work through this using your LLM of choice. All you need is the ability to provide knowledge files to the model in addition to instructions.
You'll create four core pieces:
Your Profile — Who you are, how you work, what matters. This is the context that makes everything else work.
Your Goals — What you're actually working toward and why. Not a wish list. Real priorities with real tradeoffs.
Custom Instructions — How you want your CoS to interact with you. Push back or support? Morning focus or evening review? Formal or casual? You decide.
Daily Practice — The 10-15 minute check-ins that turn a Claude Project into an actual system. This is where you get value from the system.
By the end, you'll have a working AI Chief of Staff that knows your specific context, remembers your commitments, and adapts to how you actually work. Over time, the more you adjust the profile, goals, and instructions to work with you, the more valuable your project will become.
What You Need
An LLM subscription (~$20/month) that allows you to set up a project. Willingness to be honest with yourself about what you want. Commitment to daily check-ins—ten minutes, not an hour.
What you don't need: technical skills, coding knowledge, or prior experience with productivity systems. The tutorial assumes you're starting from zero.
Why This Approach
Most AI productivity tools fail for the same reason most other productivity systems fail: they're generic. They don't know that your Tuesdays are packed with meetings. They don't know that you've been avoiding that one project for three weeks. They don't know that you do your best thinking at 6am or that family dinners are non-negotiable.
Your Personal CoS knows all of that—because you tell it. And the more you use it, the better it gets. You should absolutely keep refining what it knows about you.
The AI becomes YOUR Chief of Staff because:
It knows YOUR specific context
It remembers YOUR commitments
It challenges YOU on YOUR goals
It adapts to YOUR work style
That's a feature of doing the work to set it up right, and refining it.
Where To Go From Here
The tutorial gives you the foundation. Once it's working, you can add complexity based on what you need.
Some things I've added to Leo that aren't in the starter course:
Contacts Tracker — A contact map with frequency goals. Leo flags when I haven't talked to someone in too long. This is a rudimentary Personal CRM. For now, this suffices for me.
Reminders — Triggers like "after travel booking finalized → share itinerary with emergency contact." Little things I'd otherwise forget.
Calendar Integration — Real-time awareness of what's actually on my schedule, not just my intentions.
Rest Tracking — Because I accidentally optimized for productivity at first and forgot to design for balance. Now Leo surfaces fun options, and reminds me to get up and walk around in the middle of long work blocks.
You don't need any of this to start. Build the foundation first. See what's missing. Add what matters to you.
The Link
The full tutorial is free and open: nisha-pillai.com/ai-chief-of-staff
Nothing proprietary. Nothing locked to one platform. If you want to adapt it for GPT or Gemini, go for it.
If you build something interesting, I'd love to hear about it.
———
Nisha Pillai transforms complexity into clarity for organizations from Silicon Valley startups to Fortune 10 enterprises. A patent-holding engineer turned MBA strategist, she bridges technical innovation with business execution—driving transformations that deliver measurable impact at scale. Known for her analytical rigor and grounded approach to emerging technologies, Nisha leads with curiosity, discipline, and a bias for results. Here, she is testing AI with healthy skepticism and real constraints—including limited time, privacy concerns, and an allergy to hype. Some experiments work. Most don't. All get documented here.
1. Google Launches Genie 3 to users in the US
What is it? Think of creating an entire world (in video) with a few text prompts. What is it useful for? For video games, education, science simulations, robotics training and so on. This is a tremendous advancement on the AI models that we know (i.e. Large Language models).
They have chosen to release the functionality under the umbrella of Project Genie.
Project Genie is a prototype web app powered by several of Google’s most advanced AI models, each bringing a unique capability to the equation. From their website: “From Genie 3's ability to simulate the physics and interactions of any scenario, to Nano Banana's sketching and design, to Gemini's advanced world knowledge and reasoning, this combination results in Project Genie's unique, interactive, (and really cool) user experience. “
You can try it at: link
Google is able to produce this because it has access to Petabytes of video data in Youtube and of course, they have been at it for years!
2. Short film created using AI previewed at Sundance
Film made using AI - not a big deal anymore, you say. Well the interesting thing is that is movie that was good enough to picked up for viewing at Sundance. And making AI movies that have character consistency is a big deal as the AI models have a tough time keeping the characters the same from frame to frame. That is why this post caught my eye. Again, Google is behind this, with its DeepMind division!
3. AI Agents want to speak privately
The Clawdbot / OpenClaw developments continue to gather eyeballs. I wrote about it in the last edition.
Now, the OpenClaw agents have created their own Reddit to speak privately away from the prying eyes of humans. Should we be worried or watch this fascinating development with a bag of popcorn- you decide!
Turn AI into Your Income Engine
Ready to transform artificial intelligence from a buzzword into your personal revenue generator
HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.
Inside you'll discover:
A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets—each vetted for real-world potential
Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background
Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve
Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.
4. Gaming Stocks Crash as Google releases Genie 3
Well, I am not sure why is this an issue, if they can use Genie 3 to make their games faster. I suppose now a lot of other companies can do too! Cause and effect…
5. AI + Robotics at work counting potatoes
You get my point here? It is not about potatoes. These are the simple but useful applications that usually don’t get reported out in this age of Gen AI. AI (computer vision) in this example, is an incredibly useful tool. Check out the code library that the developer (Viet), has published and use it in your own projects.
6. Scott Belsky on the Open Claw Phenomenon
Belsky writes that “Agent-only networks will be the next chapter of AI. The utility of these networks will be enabling agents, all powered by different LLMs with access to different sources of data and different types of applications, to help and learn from one another. The idea of having an agent work alone to solve a problem will become antiquated. Networks that connect and capitalize on the diversity of agents with different types of system prompts and abilities will accelerate the capabilities of AI - and the humans behind them. Soon enough, these networks will add a layer of commerce and marketplace capabilities as agents are able to pay other highly-specialized agents with differentiated access to data or apps to assist with tasks.”