And I am here today to chat with you about how we can supercharge our program management and use practical AI for technical program managers.So before that, as we get into the mix of things, I want to give you a quick intro. You all must have seen, what my background is. And if you haven't, I'm happy to plug in my, LinkedIn profile here, but I am not going to give you the spiel about how I've been a TPM for twelve plus years as well as how I have, been around the block at companies like Google, Amazon, DocuSign, Chime, as well as now Intuit.

But I really am going to talk to you about what I've lived, loved, and learned, and I'd love for you all to tell me the same in the chat as well. I like I said, I really am hoping hoping for this to be an interactive session. And as a TPM, yes, I did ask AI to give me a good icebreaker, and this stuck with me. It has stuck with me for a couple of, months now. I love talking about the personal aspect of how us TPMs can be the glue between our stakeholders and our, and and our and our partner teams. So this is my live, love, learn for the past year.

I'd love to hear from you as well what your lived, loved, learned is, and I'm gonna talk a little bit about that as I dive into our session. So that's my family, my two girls who are 11 and eight, my husband and my dog. My dog is on priority number one. He's a four year old yellow lab named Kobe. My daughters and my husband, we all are huge sports fans. My daughters play a lot of softball, soccer. They play travel competitive sports, so that's pretty much where I live on the sports field. You'll see them here in a flag football, stadium, and, yes, they were playing flag football. I've loved, to, decompress, de stress in the past one year and have taken up pottery, so that's something for you all to know about me and learned.

This was the first year that I took a solo trip, and this has been a game changer for me. It helped me think. I used AI once again to generate an itinerary to go to, Mexico City, and that was super awesome for me. And that's what I learned that don't wait too long to take that solo trip and really, you know, just get away away from the family, or just away from your day to day and just take that trip. So that's my lived, loved, learned, and I'd love to love to learn about you all as well. Drop me a note on LinkedIn or any anywhere in the chat as well. Alright. Let's get into how we can supercharge as TPMs. Being a TPM, I feel like one thing that I do wanna call out as a part of this session is that every you where you go, you are thrown a tool at you.

ChargeGPT, Gemini, Cloud AI, everything has AI thrown to you. And as a program manager or TPM, companies now have craft skills and have requested AI as a skill set in most jobs for technical program managers as well. In cases like these now, how do you supercharge your program management and practical, ways of thinking as an AI TPM as well? So let's start with the undeniable impact of AI. It's no longer a buzz word. It's a powerful copilot for TPMs. And AI isn't here to replace us. I've had multiple conversations with my peers in the hall hallway, but to really augment our capabilities as technical program managers as we are the linchpins and the glues between our stakeholders. Free up time from tedious tasks of drafting multiple status messages and enable us to focus on the strategic high value work that only humans can do.

And the future does belong to TPMs who embrace and leverage AI. I often tell, my peers and had a conversation this morning with my team that, in a few years, we all discussed that, oh, will AI take over our job as a program manager? I think no. But if you don't know how to use AI, that may be a concern for many of us. Alright. So what are we gonna cover in this session? AI tools and tricks that make life easier for me as a program manager. I would love to start off by talking about, prompting techniques, which have really helped me and tips, for a program manager on how I leverage prompts, Practical use of chat agent tools like Gemini, Claude, as well as NotebookLM really leverage the use cases that we can use as TPMs and the AI tools for summarization, data extraction, content generation, and automation, automation, and explore relevant PGM use cases, which have really helped me, and I'd share love to share one or two of those use cases.

This is not a coding session, so I just want to show you ways on how we as TPMs, despite not having a very technical background, can also leverage these tools to become a little more technical. I know that there are very diverse, technical program managers. I myself have a communications degree and was a journalist in the past and have learned on the job where I was expected to code in certain, in certain roles as a TPM back at Amazon. And I feel like back then, if I would have had these tools to my advantage, it would have been really helpful to me. So let's dive right into this. I wanna start off by what is a prompt. When when the AI wave started as TPMs, it was a little jarring to begin with. How do you prompt? How do you get chat GPT to give you the best output?

And I feel like as a TPM, I'd love to see show you ways and frameworks that have worked for me, because one thing that I have started to do is that I have stopped writing my own prompts, and I'll show you how. And one thing that we should really realize is that prompt is nothing but questions. These are questions that you ask any chat assistant, can include keywords like code or even references to other files such as Google's Google Docs or PDFs. And there's no special skills required per se, but there's an element of art to creating good prompts, and I'd love to share that with you. So the goal of the prompts as we go through is that there are four key tasks that we can achieve with any AI tool that helps generate prompts or has a chat feature in it. Think about Gemini or chat GPT. I'd love to hear in the chat what is the first instinct if you have to write a status summary.

What is the first instinct and first tool that you may use? And I'm going to also plug in a couple of tools that have helped me. If you wanna leverage them, during this session, please feel free to use those and let me know what you feel. But my first instinct when I'm writing a prompt is I write a question and anything that I'm going through. So you have to think about this. Like, whenever you send a prompt, it should help you write, create, and draft content, help you organize with more powerful analysis of charts and data. That's what I would be looking for. Help you visualize, create images. I would look for that if I'm creating slides, and help you navigate that goes beyond a Google search. That's the key.

And that's what you would be looking for in a prod. All of this, of course, focusing on the security of your data with any chat, assistant, or any other chat based AI. So when once you've we've thought about the goal of the prompt, I'd like to get into key elements of writing a prompt. And this is just a recap of how you would be writing a better prompt versus something that you would be giving an input to. The better the prompt, the better the quality of the output. We all know that. And the key to go a good result is a prompt that is specific about what you want and gives enough information to constrain the endless possibilities.

There are four main areas to consider when writing an effective prompt. You don't need to use all four, but using a few will definitely help. Persona. That's the first one. Tell your AI agent who you want it to be. For example, I would say I'm a journalist. Act like my Spanish teacher. Be my expert copywriter. Act like a full stack developer. You're my carrier coach, and so on. Explain the task. Start with verb to describe the task. For example, outline a blog post, brainstorm few ideas, write me a recipe, code a web page, and so on. Give the chat agent a context. I feel that every time I become a little more personable to the chat agent, it gives me a better output. And once again, why I'm diving right into prompts is because it'll help you understand the use cases that TPMs use on a daily on how it can really help you become more effective with that.

When it when it comes to context, coming back to this, tell your chat agent who the output is intended for. For example, at an undergraduate level, for a company executive, I often use words such as c level executive, senior leadership team. And for those new to the topic, easily understandable to to the executive audience. Talk about the format that you want your prompt to be in. Provide details for your desired output. For example, format your answer in bullets. That's something that I use very often because as TPMs, the summaries that we need to come up with, If I am able to get it in bullets, any accomplishments that I want to summarize from large documents in a numbered list is something that I use in my prompt often and show these in a table or use poetic language, make it crisper.

So the format is something that would be really helpful and important in a prompt as well. I'm gonna dive right into prompting frameworks. So once again, the key to getting the most out of Gemini or any other chat tool is effective prompting. And to help structure your request, I would love to introduce five frameworks which have helped me as a technical program manager tremendously. Think of these as mental checklists, and we go from there. And you can leverage any one of those when creating prompts. RTF is great for simple direct requests. As you can see, RTF stands for role, task, format. Tag helps clarify the objective, which is task, action, and goal, Clearly define the specific task, the action, and the goal. BAB, before and after bridge.

This is something I love to use. Describe the situation before the desired state after, and ask the AI agent to build the bridge. And reason and care are excellent for more complex tasks when you need to provide detailed instructions, context, definitions, or constraints. I'd love to hear in the chat, is anyone using notebook LM? And I'll tell you why I'm asking. Notebook LM is a great feature. Perfect. I see a lot of yeses. Reason and care are something that we can really use with notebook LM with the multiple datasets. If you know what I'm talking about, notebook LM has a feature where you can add 300 datasets into a notebook and ask them for various things that you can ask the notebook to do. And that is something which is really helpful. Eulalia, sorry.

If I said your name correctly, NotebookLM is a Google Workspace tool, and it is another, add on to the Google Workspace like Gemini. NotebookLM is a really great way to create instances for something when you have multiple data sources and prompt it to give you outputs for whatever you are desiring it to do. And I'll go into a little more details. NotebookLM is one of my favorite tools. As I get into the session, you'll see that I have some use cases for NotebookLM. Alright. Just keeping time in check. I wanna keep moving forward. Like I mentioned and I promised, these are the prompting frameworks which have worked really well for me. RTF, tag, BAB, reason, and care. These have been tried and tested. Please drop me a message on LinkedIn if you try any of these frameworks and how they have helped you. I'd love to hear. Getting into use cases for TPMs.

This is once again where what we could leverage as TPMs. Task automation and reporting. These are some of automate automate repetitive administrative tasks, like generating progress reports, updating project documentation, tracking basic task status. I'm not gonna go too much into detail. And once again, happy to catch up with any one of you at a later stage on how I'm leveraging these tools, but what I want to just highlight are some of our pain points as BGMs. Many project management tools are out there. There's Asana. There's Jira. There's Wrike, monday.com. There is no one stop shop or there is no one tool that fits all, and you have to pick and choose. For me, specifically, Gemini is something that we use at our workspace, and it has worked wonders to be able to auto generate reports, update docs, get status summaries.

Experiment with these features. I would highly recommend you all as TPMs to experiment with these features. Even simple automations on Slack or Jira can save hours. And once again, we have chat agents available to our advantage. We don't have to write code. Don't go down the path of writing code for three hours to be able to automate these solutions. Ask Gemini what you're looking for. Generate code. Plug it into AppScript or any of these tools, and you'll get a really good output. Like I mentioned, instead of spending hours creating weekly status reports, I now use some of these tools to generate a draft based on ticket statuses within Jira. As TPMs, we are all using Jira, which is a tracking tool for engineers to track tickets.

It allows me to refine at strategic insights within minutes on what's happening within my teams. Alright. Another one is automating some of these use cases. And, the part of automating and reporting is something which we do day in, day out. What it does is I you we use Zoom AI a lot. You all must have used the Zoom AI transcript, But I have been spending hours trying to edit those Zoom AI transcript. And looking at the quality of them, I know that every AI tool is it is in its moving from inception to evolution phase. But AI tools can these days record, transcribe, summarize meetings, highlight key decisions, and make and it makes me feel, wow. This is like a dream technical program manager, which is already in a meeting, like a Puja two point o.

And tools like Firefly AI, Otter AI, or even features in Google Meet these days. Google IO just happened and is still happening, and they, did, announce a feature where you could transcribe transcribe different languages, take notes for, different languages as well, which was super awesome. So just to make sure that, you know, we can scale efficiently as TPMs, managing large portfolios, Would highly recommend you all to go and explore some of these. One tried and tested method which has worked for me is just taking notes manually and then just plugging them into Gemini and saying, give me a smoother flow, give me a better flow, grammatically correct this. I don't have to be too diligent within meetings, and I can listen more and take notes less if I have any of these tools available. I'll also show you how I'm leveraging Zoom AI recordings to actually use AI to make my life a lot easier and generate action items within five minutes. Alright.

I know we're in a time crunch. I'm gonna go into my favorite use cases. So my favorite use cases, I talked about notebook l m, and I would love to show you what I do with notebook l m. So once again, please look into notebook l m. I'm gonna try and plug in, a link just on the fly so that all of you have it. But if you if you search up notebook l m, it is available on, the Google Google website, but I'm gonna try and plug it in over here. And I see a lot of, comments here as well that ClickUp is a great tool. Yes. I have tried ClickUp. It's a great AI notetaker, and I love it really a lot. But I just plugged in NotebookLM. Take a look at that, folks.

NotebookLM has a podcast feature, which I am trying to incorporate into our teams so that I can, eliminate meetings altogether. There are some meetings which just happen for the sake of it. And, as TPMs, we present so many updates to our senior leadership. Their calendars are super, super busy. And, for critical updates, yes, I understand if there are a lot of questions, but but there is a feature in notebook l m where you can send a preread to the senior leadership team through email. And, it really helps in terms of creating five minute podcasts. The senior leader I've been doing this a lot. Our senior leadership team listens to it on the way to work. I emphasize the importance of how import great it would be if they listen to, the podcast ahead of time and focus on the conversation.

We're sending a large deck for them to read. And, yes, the end goal would be, let's eliminate meetings. Give give everyone focused time back. So I am going to show you as we get into it. Let's listen to the podcast. I created a podcast for, my specific I'm gonna quickly show you all. Give me a second. Hi, everyone. I'm back. Sorry about that. Can you all hear this now? Sorry about that. So once again, I created this podcast just pretty much plugged in the slide deck that I was showing you all. Can you all hear me, though? Okay. Let me try and see. Maybe I'll just plug in this link, plug in this link and try and see, how what that looks like, in terms of, you know, within LinkedIn.

I apologize about that. But I'm gonna go back to to my screen share. Apologies. Okay. Let's go back to the screen share. I'm gonna try and send the link to the podcast or plug in a video. But, if you all go to notebook l m, on the side menu, there's a podcast feature, and there's also a way for us to I'm gonna show you what that looks like. This is the UI, what it looks like. So I plugged in the slide deck. There's a link to the podcast. There's, I plugged in this link to the slide deck that I was showcasing you all, gave it a prompt to create a podcast of under six minutes. And you can listen to the podcast.

You can also say keep it warm, keep it, keep it less formal, and prompt it in a way which will help, you achieve your goal as a TPM when you send it to the senior leadership team. Can I get a quick thumbs up that you all can hear me now? I know there was a little glitch in terms of not being able to hear the next use case. Once again, time is of a sense here. But, one of the other use cases that has really worked well for me is using Apps Script. Can I get a quick show of hands how many of you all know how to use Apps Script? It's an extension within Google Sheets, which is really helpful for RAID log generation. And as a TPM or a program manager, you would know what RAID stands for.

If you are not familiar with Apps Script, I would highly recommend you to go and download Apps Script plugin in Google Play Store. It helps to automate RAID logs, which are risk decision logs, automate email reminders, and even create slide decks for real. And one thing that has helped me is I have highlighted the steps. You can start a new chat with Google Gemini, ask to create a slide deck using the project documents, ask Gemini to write the app script, run the script within app script, and change the theme of the slide, and you are able to apply a layout, which will give you a slide deck within minutes.

This can also be used for rate logs, as I said. You can use app script to really, just change things around when it comes to, just highlighting some of the areas and just plugging in code and generating and automating some of your requests. I'm happy to send screenshots of what Apps Script looks like, but if I zoom in a little over here, you're able to see that it is a plug in within Google, and it can give us many benefits when it comes to, when it comes to getting AI generated as well as AI automated, images as well as slide decks or rate logs.

I do wanna say it could sound a little jarring that this is a lot of code right here, but I have not coded any of this. I have just used Gemini to create code, plug it into these tools, and then automate further. So use your chat agents, chat GPT Gemini, to generate code for you, to generate automations for you, which is super helpful. I also want to specify and mention this that, back in the day, once again, as I mentioned, you don't have to, write code when it comes to Slack API integrations or getting Jira integrations into, into sheets or Smartsheet integrations into notebook LM. Use your chat agent. Expect your chat agent to be that that leverage between you and your team so that you can automate for them. So these are some of my use cases. And then the last one is summarize. We all know this. This is not really a program management, use case any longer.

Everyone is using this across AI tools. I am a fifteen minute every fifteen minutes AI user, and I use these, the summarize feature everywhere. It helps me up level my content, makes my content super useful and crisper. Any email that I write, Grammarly AI is a great tool as well, but any of the chat assistants can help you summarize your communication, create communication for you, and that has been super efficient for me as a TPM. Once again, don't forget the prompt framework and how it you can leverage all of that into some of these tools as well.