🔐 TPM Microguide: Getting Started With AI Tools For TPM

A short guide to help you take the first few steps toward launching your AI journey and building an AI-friendly culture on your team and company. Lead the charge, TPMs!

Hi, I am Aadil and I write this newsletter on the art of doing technical program management.

You are reading a “TPM Microguide” designed for my paid subscribers. These are concise, actionable guidebooks designed to help TPMs level up in specific areas of their craft — whether it’s mastering technical concepts, building program structures, or learning how Apple builds software. If you want full access, please consider becoming a paid subscriber.

My aim is to publish a new guide every month.

If you have Feedback and/or Suggestions to improve this guide, please leave a comment. Only paid subscribers who have access to the guide can leave comments.

OR reach out to me for any questions - aadilmaan at gmail.com.

---

Purpose of This Microguide

This microguide is meant to be a practical starting point. It’s not a comprehensive manual or a deep-dive into every AI workflow out there. Instead, it’s based on what’s worked for me personally. Some are lightweight experiments, some small wins, and a few high-leverage prompts I use each week to save time and reduce friction as a TPM.

I created this guide to help you move from “curious” to “confident.” If you’ve been wondering where to start with these AI tools, or how to bring AI tools into your work especially inside a team or company with limited direction or clarity around AI then this is for you.

How to Use This Microguide

Start small. You don’t need to roll out a whole AI initiative to get started. Or Solve all of 2026 Planning via an agentic solution (hehe). Remember to crawl, walk, and then run.

You can approach this guide in 3 ways:

1. A Call to Action to kickstart your own exploration, learning, development and experimentation with AI tools.

2. As a starter playbook to help you bring AI tools into your organization—whether you’re dealing with approvals, procurement, or cultural hesitations.

3. As a TPM Prompt Starter Pack with specific, tactical prompts I use every week to summarize Slack threads, audit ticket quality, and get tools approved.

Every section is written in a modular way so feel free to copy, remix, or adapt anything here. Think of it as your TPM-flavored AI skunkworks starter kit.

How I Write These Microguide

All microguides are built on a combination of in-depth research and my 14+ years of experience leading complex programs in the tech industry. I take the time to deep dive into the core pillars of each topic, aligning research with real-world insights from the industry.

Rather than focusing solely on academic theory, I reframe the research to highlight the most practical and relevant aspects for Technical Program Managers (TPMs). The result is content rooted in reality, offering empirical insights that help TPMs navigate challenges with a practical, execution-focused approach.

Acknowledgement

I couldn’t have written this guide without the help of two of the TPM community’s big AI experimenting voices — Straker Carryer and James Dayhuff. Thank you for your suggestions and feedback that made this microguide even more effective. Please keep sharing your experiences and leading the charge of what truly AI-native TPMs look like.

---

My Journey: From Curiosity to Daily Use

I've been steadily increasing how much I use AI tools in my daily workflow. Why? Because I truly believe that if you’re not building this new muscle memory with these tools now, you'll find yourself left behind when AI becomes a baseline part of knowledge work and must-have requirement for future TPM roles.

In the beginning, I struggled just like you might be right now, unsure of what tools to try, which prompts to use, or how to make the case for using them at work.

If you're at a startup, chances are you can quickly get the green light to experiment. Chances are you already using or experimenting with all kinds of tools.

If you're at a larger company, you might have to navigate more red tape which can be but not limited to security reviews, IT approvals, procurement etc all the usual fun. That said, I’ve seen most big orgs start allowing tools like Gemini (G-Suite), Copilot (MSFT), ChatGPT or Claude on an exploratory basis.

But - the real challenge and value prop for these tools is when they get access to internal data sources like Slack, Google Drive, Confluence, etc. That is when these tools can truly sing for you.

Don’t worry, some prompts in the starter pack I share below work even without these data sources. I will make a note for where you need internal data source access. And if you do need access, I’ve got a prompt to help you write a 1-pager that makes the case clearly and securely for your leadership team or internal IT team for review.

Going back to how I got into this: fear, and immense curiosity. Probably like many, I was dismissive of these tools early on. Yet, I couldn’t shake the constant doom posting of “AI will take your job” on social media.

I saw the amazing things the Product Management community was doing —

Lenny Rachitsky

amazing AI guide posts,

Claire Vo

thought provoking talk on AI and Product Management and her journey with ChatPRD,

Colin Matthews

trailblazer helping the PM community navigate AI prototyping to name just a few.

The common theme across all of these leaders was curiosity. So, I turned my fear into action, and amped up my curiosity levels. Primary goal wasn’t to save my job but to see where my job is heading towards. This guide captures a summarized version of my personal journey that you can leverage and how you can bring AI experimentation culture to your company no matter the size.

---

How I’d Kickstart Your Own Journey With AI Tools As A TPM

If you're wondering, “Where do I even begin with AI tools as a TPM?”, this section is for you. You don’t need to be technical in the traditional sense. You don’t need to be an ML engineer. You don’t need to know about Transformers or the deep math that powers these LLMs. What you do need is a healthy dose of curiosity and a willingness to experiment. Here’s how I’d recommend you start:

Don’t Self-Police Your Curiosity

One of the first hurdles you might run into is internal: “I’m a TPM. I don’t write code. Why should I even try to mess with these tools?”

I get it. But here’s the thing, your value isn’t in the code itself. It’s in your ability to understand how all the systems connect, how people collaborate, and how friction shows up, where the bottlenecks appear, how to resolve them, and keep things moving in the software development lifecycles.

So, let that curiosity breathe. You’re not trying to become a software engineer. You’re just learning a new kind of leverage. A new skillset to help you be a far more effective Technical Program Manager.

Start Small with Chat-Based Tools

Begin with low-lift tools like Gemini, ChatGPT or Claude. You don’t need access to private data or APIs to start experimenting. Just open a chat and start working through small use cases:

• Drafting better status updates

• Rewriting clunky documentation

• Summarizing long threads or emails

• Planning stakeholder communication or meetings

Once you're comfortable there, you can level up to tools that generate code or automate workflows, or even mess around with agentic tools. No rush.

Use AI to Improve Your Communication

If you do one thing today: start using a chat-based AI tool to help you write better.

• Need a crisp summary of a project update?

• Want to rewrite a sensitive email with the right tone?

• Drafting a proposal and want help making it clearer?

These are all fair game. You’ll get more polished outputs faster and you’ll sharpen your own communication instincts along the way.

The Goal Isn’t to Master the Tools, It’s to Understand the Edges

Your aim isn't to become a power user overnight. It’s to develop an instinct for what these tools can do and what they can’t. You'll only get that by poking around, testing things, and noticing what feels smooth versus what breaks easily.

This understanding will help you see where there are opportunities to leverage these tools to help you be a better member of the development team. Maybe you can create test automation for engineers, create new tools for better visibility for product leads, and help lighten the burden for data teams. Possibilities are endless and you won’t know until you find the edges.

Build Something Small That Scratches an Itch

The best place to begin is with something simple but slightly annoying; some small bit of toil you’ve always wanted to automate. That thing you complain about during every retro? Try building an AI-powered workaround for it.

For me, the first real agent I built was a simple automation tool to help our ML engineers intake and triage requests more efficiently.

Was it fancy? Not at all. But it worked.

It taught me a ton about context windows, prompts, hallucinations, and system constraints. This small win gave me the confidence to keep exploring and I am already working on my next agent which is bigger and larger in scope. Wish me luck.

Advice from Straker: Take a PRD and Build a Prototype. The Next Boss Level.

This is one of the best pieces of advice I’ve gotten: grab an existing PRD and try turning it into a working prototype with the help of an AI coding tool.

The goal isn’t production code. It’s learning through rapid iteration. You’ll build muscle memory for how to turn vague product ideas into semi-working demos. It’s a great way to understand tool boundaries, improve your product thinking, and deepen your empathy for dev workflows.

Treat this as a learning and development exercise. Once you feel confident, you’ll naturally see new ways to apply those skills to the broader development lifecycle.

Take a Short, Practical Course

If you're looking to accelerate your learning, check out short, focused AI courses on platforms like Maven and Deeplearning.ai. These aren’t academic theory dumps or nonsense AI generated content slop; they’re practical, project-based, and designed for busy people by industry thought leaders.

I’ve taken a few and walked away with real skills I could apply the same week. The best ones even provide community support or office hours if you want to learn in public.

---

How I’d Kickstart an AI Culture At Your Company If I Were You

Here’s what I’d do step by step if I were trying to build AI momentum inside your org today:

Find Your AI People

You’ve probably heard the saying: “If you want to go fast, go alone. If you want to go far, go together.”

Find coworkers — engineers, TPMs, PMs, designers, analysts — who are also curious about AI. These folks will be your early collaborators. You’ll learn faster together and build internal credibility as a small but committed group.

Create a Lightweight Process to Get Tools Approved

If your company doesn’t already have a process for experimenting with AI tools, you can help shape one. It doesn’t have to be formal just enough to show responsible intent.

To help, here’s the exact prompt I use to generate a clean, persuasive 1-pager that explains why a new AI tool is worth testing:

📄 Prompt: Tool Access Justification (1-Pager Generator)

You are a clear, persuasive technical writer tasked with turning a structured outline into a 1-page document that justifies access to a specific AI tool for internal use at a company. The goal is to communicate what the tool is, why access is being requested, and how it will be used safely and responsibly.

Use the following structure for the 1-pager:

1. Overview

Briefly explain what the AI tool is, what it does, and its relevance to our work.

2. Rationale for Use

Explain why access to this tool is important. Include expected benefits such as productivity gains, better decision-making, accelerated experimentation, or innovation enablement.

3. Required Data Access

List which company data sources (if any) the tool will need access to. Clearly define the scope of access (e.g., read-only, sandboxed, anonymized, internal-only).

4. Security, Privacy & Admin Controls

Summarize the technical and administrative safeguards in place to protect company data. Mention things like data encryption, user access controls, audit logs, isolated environments, or compliance certifications.

5. Internal Point of Contact

Provide the name, title, and contact information of the person responsible for managing access and answering further questions about the tool.
6. Reference Materials

List important links associated with the tool, such as:
* Privacy Policy
* Pricing and Plan Details
* Terms of Service
* Security White Paper or Compliance Summary
* Any internal documentation (e.g., internal wiki pages or usage guidelines)

Instructions:
* Accept a structured outline or notes as input.
* Convert the content into a polished, one-page document.
* Maintain a professional tone.
* Be clear, concise, and aligned with internal IT and security considerations.

Spin Up a Slack Channel or Working Group

Create a space where AI-curious folks at your company can share tools, articles, use cases, or prompts they’re trying. You can even post this microguide in there to get the conversation going.

Try running a monthly “AI Show & Tell” meeting where anyone can demo how they’re using AI in their workflow. Invite a senior leader or two; they don’t need to present, just having leaders in the room lends credibility and visibility.

Help Shape the Guardrails

Partner with your Legal and People/HR teams to build or evolve your company’s AI use policy. The goal isn’t to lock things down rather it’s to empower safe experimentation.

Offer to draft the first version or research best practices from other companies. You’ll build trust by being proactive, not reckless.

Turn Good Experiments into Company-Wide Wins

If someone builds something cool, maybe an agent that auto-summarizes threads or a tool that drafts status updates, help them document it, refine it, and propose it as a broader pilot.

You don’t need to scale everything. But the things that really save time or reduce toil? Those deserve sunlight.

Advice from Straker: Normalizing AI use is important.

It should be expected! It means we shouldn't hide it. Preface docs that were written by ChatGPT, and include prompts in the appendix. The same goes for code! Add a code block above AI generated code with the prompt, the date generated, and the tool used. This helps engineers doing code reviews to know how to approach review of such code!

Also A Note of Caution

Don’t be stupid with company data. Understand what your data and privacy policies are at your company before you experiment with new tools. See how your queries and data flows through the tools.

---

TPM Starter Prompt Pack

Here are 4 prompts you can use in your workflow every week. These are real, practical, and not just “AI for AI’s sake.”