Thanks so much to Mike Chambers (AWS ML Hero)for inviting me to chat about key takeaways from AWS re:Invent 2021!
We both live in Brisbane, Australia so we were both unable to attend re:Invent in Las Vegas in person, but we still found a way to experience the community aspect of the event!
Watch the video here:
You can follow Mike on Twitter here: https://twitter.com/mikegchambers
And if you're interested in learning more about Machine Learning on AWS, I really recommend Mike's course for the AWS Machine Learning Specialty Certification aws-mls-c01 Exam!
Mike: Hello everybody, well Reinvent 2021 is still going on in Vegas. I think around about now it's party time in Vegas and they're all partying it up at replay. But for some of us who aren't able to be there, well, we can't be there. But reinvent for me is all about meeting people and connecting with the community. That's why I go to reinvent. And so I'm doing that in my town and I was really grateful to find Brooke who is community builder for AWS in Brisbane, my hometown. Hey Brooke!
Brooke: Yeah hi, thanks for you having me! I would add that it's not just that people are partying after reinvent, now they're probably all starting to get sick, not with the panini, but with probably any other thing that they've come down with from partying too hard. So thoughts for everyone currently struggling if you’re watching this
Mike: Fatigue? Yes, yeah. So this is the first time we've met, like we've had like literally two minutes before I turn the camera on. But yeah, it's a pleasure to meet you. I like walking down the corridors at reinvent, bumping into people, and so this is what we've done virtually. Real really kind of, virtually speaking, so how's reinvent? Been for you on this side of the world?
Brooke: Yeah, very chaotic. I'm no stranger to bad time zone differences. I was actually supposed to move to America in April 2020. So I worked for most of 2020 in both Brisbane and New York Time zones, which were about 14 hours difference, so I knew ahead of time that I wouldn't be able to do that this week 'cause I had so much of my actual normal day job to do, so I'm Head of Enablement for AI, Machine Learning and Data at Blackbook.ai , which is an AWS Partner as well. So doing that all around everything. Constantly jealous not only of everyone having a wonderful time at reinvent, but all of the time that they had to actually dedicate to watching things. I think I'll be catching up for about a month after this at this rate.
Mike: Yes, yeah. Well so for me, going to reinvent. I've said this before. I don't think this is a secret I've never been to a reinvent session. Any of the reinvents I've been to. Have you been to reinvent?
Brooke: I have not been to any reinvent sessions or any reinvents.
Mike: So well, having been there, I've not been to any other sessions either. I would just catch up with them afterwards. Online on demand, even pre covid times, but it is about the people and yeah no so obviously. Hopefully next year we can both be there. Yeah, and connecting with those people. So machine learning as you said, so we both sort of share that common interest in machine learning. Anything in particular that sort of caught your eye in relation to a machine learning at reinvent this year?
Brooke: Lots of different applications of things. I work in very Applied AI stances, so although I studied pure maths at uni but now it's very very applied. So how does this connect to industry? And I think as part of the maturity of lots of the services, the way they're tying in with different things is really good. So for example, the chat bot builder for Connect. Connect which is my third favourite AWS product. So now being able to use those call transcripts to make a chat bot because you could always make transcripts and you could always make a chat bot, but for many organisations that leap was just. Too hard, so that's really cool, and even things like amplify the front-end tools, not a front end developer, but I know a lot about Figma so I can see that getting used at scale and really large organisations as well.
Mike: That seems to me almost like a bit of a theme reinvent was wrapping easier tools on top of existing services, so making, lowering the barrier to entry for a whole bunch of things in tech.
Brooke: Yeah, it's a different view of accessibility, so obviously accessibility for actual accessibility is really important, but I think. As the underlying tools are getting more mature, people are now able to ask those applied questions about how does this actually fit in with what we're doing? Or how can I actually use this? Because especially for pretty much all of the AI and machine learning services, they're very good back end tools, but it's not as tangible as something you can just work out yourself. So I think having friendlier front ends for things is a really good trend to see happening, and that's something that I'm hoping will continue as everything continues to just mature, and I imagine lots of this serverless aspects of things will help, 'cause that's a really big maturity step as well. So I think yeah, especially next year after that it'll be even.
Mike: Sure a couple of things from what you just said there. That sort of tie directly into announcements, so different frontends, uncomplicated tools. So canvas. Canvas is, I think, as we've discovered on live streams on this channel before as well, live Canvas is a different front end onto SageMaker studio, like it is the same tool. That's great 'cause it really does sort of drop that barrier to entry. Makes it possible for someone who's not a machine learning centric person, but is a domain expert in data. I would imagine you'd need to know what the data is you're looking at, but can make it easier for you to train models.
Brooke: Yeah, and I always talk about context and I think that's the most important part of applying machine learning to business things. Is understanding all those contextual elements and how it all fits together. Even I think you can add automatically. I saw in Allie’s [Allie Miller] part of the keynote about adding the holidays in. That's really good context that before people just wouldn't have thought of because it would have been too hard. Yes, I think that part of Canvas is really good. Also, I was joking on Twitter, but um, the illustrations they have of the girl and the things I like her outfits and I aspire to make I don't know. Hopefully that will be a side merch line of just I can go in Halloween costume is the canvas girl next year.
Mike: This is the Etsy store that you're going to set up, right?
Brooke: Yes, OK, manifesting itself in jumpsuits so I can feel like the canvas girl!
Mike: So something else you mentioned just in that bit before as well, was just a serverless bit, and they announced a serverless inference end points for its age maker, but they almost set it as though it didn't really matter, and they kind of just moved on. Yeah, that to me was really exciting. What is that? Something that you think you might using? You do a lot of the practical business side and actually seeing the rubber hit the road in terms of machine learning as well. So yeah, do you see a big use case for that?
Brooke: Yeah, definitely, especially when that ties into all of the sustainability objectives. I think as part of I don't know, the serverless part is good for people to actually using it, but when you look at how that underpins all of the other sustainability pieces of the reinvent puzzle, I think that's going to be really helpful as well. Just now, people are really starting to strive for that more, and it's getting reported on more Boards and Company Director levels, so anything that enables that process alongside other technical maturity in the organisation will really help them.
Mike: Not machine learning, but well, not directly machine learning, but another thing that was announced, just very recently was the. At sixth pillar of the well architected framework, which is sustainability which I just have to mention. Obviously because you just talked about serverless in the sustainability effort there. I’m super excited about this. 'cause the idea that there is now this wealth of information that will support businesses to actually have real tangible conversations about sustainability rather than just saying yes, we think we're gonna do the right thing in 20 years time. There's something that people can actually do now.
Brooke: Yeah, and I started. I've been researching this heavily for the last six months or so and I started working out how I could manually. Calculate a lot of these things and then I saw that reinvent was coming up and I was just like crossing my fingers. That 'cause I knew I was like oh so I can get this from this and this from this and plug it together. And then I was like hang on, there's going to be some sort of announcement I saw there was a state sustainability guide for reinvent. And then I was just hoping fingers crossed that they would release the more mature features. And I got so much more than that with their well architected pillar. So I think that's good because it will force people to think about it and take it into account. Even so, because even if they're not trying to improve things, you can't improve on something you don't understand the baseline of. So I think a huge piece of this will just be people understanding where they actually are sitting as a baseline, because there's no coverage of that at so many different organisations.
Mike: Yeah, we're really supporting that, and the other part to that. Just before was it the day before? I think it was in Peter’s Keynote was the Dashboarding and the metrics around carbon emissions. All that kind of stuff. And that goes to your point as well. That just supports you. And I was just having a conversation just before I came here and. This will be up on the channel at some point as well with Luke Hargreaves, who was one of the principal architects behind actually putting together a lot of this stuff and so that was a really interesting conversation. Part of that was I think it was my suggestion and if it wasn't I'm going to take it as my suggestion that we actually develop like a little open source widget that can go on companies websites that can hook into those metrics so that they can just be here we go. This is this is how good or bad we are. That would be cool.
Brooke: Yeah, if you've ever used like in your going Chrome developer tools and you can look at the lighthouse audit for websites, it gives you four different scores that are traffic lights coloured as well as one to 100 for performance and accessibility and SEO things like that. But it's really good. Very assigned to this, but it's really good if you are running for example in E Commerce store and you want to see what your competitors stores are like relative to yours in terms of performance, but it's just in Chrome developer tools and you can look at how everything is working and the idea that it's sort of scores you on these. Different levels, something like that, I think would be really cool. Just because it not only gives a baseline for you and then you can map any improvements you have, but you can also then see how you're comparing to others, and I think. People are really fuelled by competitiveness, even if they're not fuelled by just being better for the planet. So I'll take whatever I can get for this point. Whatever pushes it forward. Yeah, absolutely.
Mike: Alright, so so. Just before we started rolling here, we were just talking about reinvent what we took away from it. You have an additional thing that you think that people can sort of take away from reinvent.
Brooke: Yeah, yeah, so I do lots of speaking about speaking, which is incredibly funny when I say it like that, but lots of things about how to present your data so it's persuasive. How to make your voice heard just because as a young woman in tech, that's not always easy out of the gate. So I try to make it easier for other people. And I mentioned this in my reinvent session guide, but I think it would be really helpful for anyone in tech. Once you've had your watch through of all the keynotes, the exciting time when you're excited about the new features, watch it again through the lens of a business person and see how they're making the points, see how they're communicating things, because it is an absolute masterclass in tech communication. I don't know how big the teams are that are writing these talks, but it's not Swami writing his own machine learning keynote.
Mike: I’m sure he’s involved, but probably there's quite a lot of other people.
Brooke: There would be a lot of people helping him with that, and obviously it would be fantastic if I had that every time I do a talk, but that's not the case, so make note of how they're making points, even how they transition between some of the business, technical case studies and then the technical underpinnings of that. You won't get a better overview of that in action, so I think it's really valuable to you if you can really work out how that's actually happening, and look at how they're communicating and even looking at what's on their slides and how they communicate those slides. It's one thing to take that away from more business presentations, but they're presenting about AWS services. They're presenting about data, they're showing data visualisation. Look at that and see what they've actually done and what stands out to you so that you can use that. Or take what you want from it next time you're doing a talk or next time you're presenting to the board, and I think it would just help so much to be able to communicate lots of this to business people, whether that's for if you're trying to sell them something, or if you're wanting to be able to transition to one of these new services internally and you need buy-in, fantastic there is someone that's showing you how to communicate that to a business audience using benefits that will make sense to them. So think about how you can use that. Communication piece,
Mike: I like that a lot, I actually watched on an inconvenient truth, which was many many, many years ago now and strangely on topic still. But that was one of the things that I sort of wanted to take away from that I actually demonstrably changed my PowerPoint presentations because of watching that, and they did also have a massive team of people putting that presentation together. I think that worked from Apple. They had a big keynote steep team working on it, but just don't put. Loads and loads of bullet points in your presentations, but also to your point. Listen to the way they were talking about. Actually, the way that they sliced up the message is that?
Brooke: yeah, you can transcribe them. Obviously, in AWS Transcribe or if you are lazy or efficient, you can just in YouTube. You can open up the transcript to the side and just copy paste it and even looking at different you can do text analytics on it. I went through as a. Like a starting point, I just count how many times they talk about various things. So how they talk about how many times they bring up data? Or you can get a lot of sense of what's most popular overall in a talk. If you're looking at different service names or when they come up, but then even if there is something in the transcript that you think is a really powerful way of communicating, you have it in the transcript so you can clip it out and put it in notes or something and it will. Just it's a weird way to interact with content if you're not used to it, but it's really helpful in working out how you can level up from it. Just because if you tried to watch a course about how to present technical content. There's no way they would be able to put everything in that was specific to AWS. Sure, it would make sense, but just approach it like that and then get more value out of it. That way.
Mike: I like that. So go watch all the presentations that you want to watch and learn all about the services you want to watch. But then consider going back and looking about it again and sort of getting that extra layer detail from it. I really like that.
Brooke: Especially as well. Just as another side note, if there's ever any of the workshops, or if they put any of the other ones on that level 100, so it's normally the really basic level. Go and watch those as well. As if it's through the lens of someone you manage or someone that's new to tech and then look at how they present those topics as well. Because then next time someone comes to you and says where do I get started with machine learning or something like that? You'll understand the direction that they step through those points and how they sort of package things together and also where they stop. Think when you get really in deep with SageMaker and things like that. It can be hard to look at what's actually included in a beginners level. So yeah, I would say put your business hat on and then put your either entry level developer or someone that you manage. Put their hat on as well and then re watch some of the basic content.
Mike: Try and remember what it was like for you when you first started. Really try 'cause like there was that day when someone said I haven't. I haven't notebook and you go. What do you talking about and then then everything from there started to unfold but just try and put yourself back in the shoes if you before you knew any of that stuff it's hard to do. It's hard to do.
Brooke: Or the first time you heard the word container. Yeah, yeah.
Mike: Or Serverless, or any of these things?
Brooke: Absolutely yeah. So look at it through those lenses because they honestly the teams that it'll be us have been working on this for so long. And then the really big Sprint to the end. So I think you'll get lots more out of it. More broadly for your career than just the technical underpinnings.
Mike: Excellent thank you for that. So Brooke if people want to find you and connect with you and do all the socially kinds of things, where should they go?
Brooke: It depends what you want! So I have a Twitter that's just my name Brooke Jameson that's mostly AWS tech content. I was supposed not on LinkedIn about anytime I'm going to be speaking either in person or virtual events. So that's my speaking stuff over my LinkedIn. I also started at TikTok and if I talk about it here that means I have to keep posting on it, but my TikTok is all about how to transition into tech and getting a job with AWS or getting a job using AWS or just getting more into cloud for career purposes. That's sort of yeah, how they split up.
Mike: Well, thank you very much Brooke. I'm glad I bumped into you reinvent, even though we're not actually there
Brooke: Our fake reinvent!
Mike: In a very, very quiet room compared to anything that you might find in Vegas. But thank you so much!
Brooke: Thanks. Maybe you'll find a 10-year jacket under the table or something.
About the Author: Brooke Jamieson is the Head of Enablement - AI/ML and Data at Blackbook.ai, an Australian consulting firm specialising in AI, Automation, DataOps and Digital.
Learn more about Blackbook.ai here and learn more about Brooke here.