Episode 63: [Value Boost] 3 Affordable AI Tools Every Data Scientist Needs
Download MP3[00:00:00] Dr Genevieve Hayes: Hello and welcome to your Value Boost from Value Driven Data Science, the podcast that helps data scientists transform their technical expertise into tangible business value, career autonomy, and financial reward. I'm Dr. Genevieve Hayes, and I'm here again with Heidi Araya. CEO and chief AI consultant of AI automation agency, BrightLogic, to turbocharge your data science career in less time than it takes to run a simple query.
[00:00:33] In today's episode, we'll be exploring the most valuable AI tools for creating value for small businesses. Welcome back, Heidi.
[00:00:43] Heidi Araya: Thank you so much. It's great to be here with you again.
[00:00:45] Dr Genevieve Hayes: In our previous discussion, one of the points that we raised was that smaller organizations are often constrained when it comes to implementing data science and AI solutions by the cost of the tools involved. Enterprise grade tools often come with enterprise grade price tags, which can make it very difficult for smaller organizations to achieve an acceptable ROI.
[00:01:11] With this in mind If the data scientists who want to serve smaller organizations with limited budgets and resources, what's one AI tool that you find consistently delivers outsized value for smaller businesses with minimal investment?
[00:01:32] Heidi Araya: The top on my list is definitely Claude. Claude is a lesser known competitor to ChatGPT, but I think that Claude is smarter. Claude delivers better legal contracts for me when I need them gives me better feedback. Anywhere. So if you have data, you can.
[00:01:47] And so, therefore, you probably don't need to building custom solutions. If you have Claude and a little bit of data because Claude can help you analyze it. So I use Claude day in day out for all kinds of things. And I also find that small businesses are not aware of Clawed, even though they have leveraged chat to BT and heard of chat to BT.
[00:02:06] So, Today in a morning workshop, I introduced people to Claude and I had them work through some actual business challenges and they were just like, wow, I didn't even know Claude exists. I actually like Claude better than ChatGPT. So, Claude, if you haven't tried, I would definitely encourage anyone to try and even maybe play ChatGPT and Claude off each other and see which one you like best.
[00:02:27] Dr Genevieve Hayes: Yeah, I had a Claude versus ChatGPT deathmatch about six months ago I was using both Claude and ChatGPT for everything I wanted to do with them and put the same questions in and compared the answers. And it only took me about two or three days to conclude Claude is so much better than ChatGPT.
[00:02:46] Heidi Araya: Yeah, the writing is better. It sounds more natural. It's just somehow I can't express it. In other words, other than it seems smarter.
[00:02:53] Dr Genevieve Hayes: Yeah, my favorite use case for Claude is if you've got a difficult meeting that you've got to go into, use Claude to do a role play before it. And if you tell Claude a little bit about the personality of the person, it is like you are actually talking to that person and you can experiment with different ways you can take the conversation.
[00:03:14] It is so awesome.
[00:03:16] Heidi Araya: I haven't used it for that particular use case, but I have used it for other kinds of conversations on role plays. And so I take my transcripts from my sales calls and I can upload it and then ask it. Say, how could I have improved the sales call or what kind of, or even imagine there's another call coming up and I have three or four calls that I have the transcriptions for tell me, how do I close this call?
[00:03:38] What are the likely objections? What are the concerns and just have it act as a sales coach for me to prepare for my call. So you can even do that for customer discovery sessions. Also.
[00:03:47] Dr Genevieve Hayes: Yeah, I agree with the sales coach thing. I often use it to coach me on how to do things better with my business.
[00:03:53] Heidi Araya: Yeah, it's great use case.
[00:03:55] Dr Genevieve Hayes: Yeah, and you mentioned a couple of other tools that you found valuable before we started recording.
[00:04:02] Heidi Araya: Yes. So the other one that I help people understand exists is perplexity. So perplexity is interesting. Their mission is actually to replace Google search altogether. I love perplexity because I can ask it anything that I would have asked to Google search and it comes up with a nice little paragraph.
[00:04:19] And it lists the sources below, so I can click on them. So you always want to check the sources, of course, to make sure that it's not hallucinating. But I find that I have clients who want to do some research and that perplexity can help them do research 1 particular roofing. Company owner said to me, Heidi you know, I want to figure out where the worst damage of the hurricane was.
[00:04:45] So I can go there. The recent hurricane we had and I told him, well, perplexity is the perfect use case. So we went in there and he queried it and it gave him a bunch of information. He's like, well, now I know where to go as a roofer. I can find these clients here. So I do like that. Unlike chat and Claude, it gives you the actual sources of the information.
[00:05:03] So you can check in just make sure that. Yeah. You know, how legitimate that sources and that you want to trust it as a source.
[00:05:11] Dr Genevieve Hayes: I've used perplexity once or twice, but I mostly use Claude. When would you use perplexity as opposed to Claude?
[00:05:19] Heidi Araya: That's an interesting question. So Claude doesn't have real time Internet access. So what you're asking it, it's just going to give you from its knowledge base and perplexity does have real time Internet access and it's going to give you the sources that it has. As well, but in addition to that, I'm going to use perplexing more like a Google search and not to say, help me write a legal contract.
[00:05:41] So there's a bit of a different use case there where Claude, you can pretty much act as your general companion for anything from writing emails, giving you feedback as a coach, but for perplexity, I would use it to. Say, I'm struggling to figure out why my coffee maker isn't working or something that you might do a Google search, or I have this particular problem.
[00:06:03] What would you suggest? And then it will give you the sources of information. I believe I asked it. How do I get a driver's license in North Carolina? Or how do I change my registration? And it comes up there with all of the links to all the pertinent places that you need to go to change all that information.
[00:06:17] And that's something that Claude couldn't do. So again, each one has its use case.
[00:06:21] Dr Genevieve Hayes: Oh, so it might be my television stopped working, how do I troubleshoot it? And you might use perplexity for that. I get that.
[00:06:29] Heidi Araya: Yeah,
[00:06:30] Dr Genevieve Hayes: Okay. And because you're getting the sources, if perplexity doesn't give you enough detail, you could just click on a source and have a look at that.
[00:06:39] Heidi Araya: yes, exactly. And I have done that. I have had to click in more and read the actual full documentation. So perplexity does that for you, whereas a cloud or chat GPT may not do that for you. I have used both Claude and chat for troubleshooting, but I'm always like, where did this information come from?
[00:06:55] Do I really trust it? Whereas I know with perplexity, I can just go. Oh, yeah, that's the user manual from the actual manufacturer site. So I can pretty much trust that that's where that's coming from.
[00:07:03] Dr Genevieve Hayes: I get it. So perplexity you'd use when accuracy is essential, whereas Claude you might use where you want a greater amount of creativity.
[00:07:14] Heidi Araya: I think so. Yeah. Yeah. I mean, the interface is just very different too, right? If you ask a simple question, it will come up with follow up questions as well that are related to that 1. so I think the best thing is for people to just try it out. And instead of using a Google search, just try a perplexity search and see kind of what happens there.
[00:07:32] I did use it for a test case for a, let's say we're going to take a family vacation, and we want to spend 10, 000 or less, and it's four people, or what are some places that we might want to go, and then it came up with some ideas there and now you can click in and see where it came up with those ideas and why it might have suggested this.
[00:07:49] Dr Genevieve Hayes: Okay, I'm going to give this another go now.
[00:07:51] Heidi Araya: Okay,
[00:07:52] Dr Genevieve Hayes: And any other tools that you'd recommend?
[00:07:56] Heidi Araya: the last 1 that I use all the time is fireflies. It's an AI note taker, and it sits in all my meetings. And because the meetings that we have our fuel and their conversations, it generates a transcript, of course, and a summary, but I then download them and I can run them through an automation to generate a summary that comes from my own email to my clients, or I can then leverage it in any way, shape or form.
[00:08:18] So I have leveraged it to generate. So if you're not leveraging and recording your meetings using a note taker like fireflies, I think you're also missing out on great opportunity to query your conversation for whatever need you may have now or in the future.
[00:08:38] Dr Genevieve Hayes: And you could couple that with Claude by taking your notes and then feeding them into that to do all those things that you just mentioned.
[00:08:47] Heidi Araya: So I have a Claude project for each client, and so I can upload, especially when we have multiple conversations that say, I'm trying to close this project. So we've had four or five conversations already. It's a big project for this client. And I upload all the conversations and I just say, just give me a summary of whatever it was.
[00:09:05] If it's been a few weeks, where are we now? Help me close this next deal. I upload the latest one and I say, you know, summarize it for me. What are the objections? So now I have this database that I can then ask it anything. So even if three weeks have gone by and I need to just ask, what's the status or what are the next thing that's happening?
[00:09:22] It will be there with those conversations. So I never have to wonder where do we leave off? I always know.
[00:09:28] Dr Genevieve Hayes: So for listeners who want to start implementing what we've discussed today what's the first step they could take tomorrow to begin using this toolkit to help serve their small businesses with AI solutions?
[00:09:39] Heidi Araya: So, if you don't have a Fireflies app, first, I would download it either to your phone or if you're on a. Video call with your clients, because that way, from your very first conversation, you're going to have context of what problems your customers are having. And over, say, a few months of conversations, you're going to be understanding common patterns of problems that they are experiencing.
[00:10:00] So I would say first get fireflies, then you can start leveraging them by chatting with Claude. And then of course, you can back it up by doing your research and perplexity. That's how I would get started.
[00:10:11] Dr Genevieve Hayes: Okay. And that's a wrap for today's value boost. But if you want more insights from Heidi, you're in luck. We've got a longer episode with Heidi, where she shared her experiences in using AI to create value for businesses that are traditionally considered too small for AI. And it's packed with no nonsense advice for turning your data skills into serious clout, cash, and career freedom.
[00:10:38] You can find it now wherever you found this episode, or at your favorite podcast platform. Thanks for joining me again, Heidi.
[00:10:46] Heidi Araya: Thank you so much. It's been a pleasure to be on this podcast with you.
[00:10:50] Dr Genevieve Hayes: And for those in the audience, thanks for listening. I'm Dr. Genevieve Hayes, and this has been value driven data science.
Creators and Guests
![Episode 63: [Value Boost] 3 Affordable AI Tools Every Data Scientist Needs](https://img.transistor.fm/wWoN5Gw7dh417L26V-td4FwEtfm7WQbK8BfIdzFRJtU/rs:fill:800:800:1/q:60/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS80YzU3/MWI0ZjdlNDA1OTYz/ZGE1OWEyYTEyODk4/NDQ2NC5qcGc.webp)