![\"photo](\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/02\/Research_Podcast_Bhattacharya_Site_1400x788-1024x576.png\")
<\/p>\n<\/p>\n
In a world where productivity is paramount and only a handful of people have personal assistants, many of us are frustrated by the amount of time we spend in meetings, and worse, the amount of time we spend planning, scheduling and rescheduling those meetings! Fortunately, Dr. Pamela Bhattacharya<\/a>, a Principal Applied Scientist in Microsoft\u2019s Outlook group, wants to turn your email into your own personal assistant. And a smart one at that!<\/p>\n Today, Dr. Bhattacharya tells us all about Scheduler, Microsoft\u2019s virtual personal assistant, and how her team is using machine learning to put the \u201cI\u201d in intelligent PDAs. She also talks about how understanding different levels of automation can help us set the right expectations for our experience with AI, and explains how, in the workplace of the future, we might actually achieve more by doing less.<\/p>\n Pamela Bhattacharya:\u00a0Doodle did a survey of more than, I believe, fifteen hundred professionals, and what they found is, on average, people in these roles spend more than five hours in scheduling meetings, just back and forth in trying to find the right time. Imagine, five hours, every person, spending every week, to schedule meetings!<\/p>\n Host:\u00a0<\/b>You\u2019re listening to the Microsoft Research Podcast, a show that brings you closer to the cutting-edge of technology research and the scientists behind it. I\u2019m your host, Gretchen Huizinga.<\/b><\/p>\n Host:\u00a0<\/b>In a world where productivity is paramount and only a handful of people have personal assistants, many of us are frustrated by the amount of time we spend in meetings, and worse, the amount time we spend planning, scheduling and rescheduling those meetings! Fortunately, Dr. Pamela Bhattacharya, a Principal Applied Scientist in Microsoft\u2019s Outlook group, wants to turn your email into your own personal assistant. And a smart one at that!<\/b><\/p>\n Today, Dr. Bhattacharya tells us all about Scheduler, Microsoft\u2019s virtual personal assistant, and how her team is using machine learning to put the \u201cI\u201d in intelligent PDAs. She also talks about how understanding different levels of automation can help us set the right expectations for our experience with AI, and explains how, in the workplace of the future, we might actually achieve more by doing less. That and much more on today\u2019s episode of the Microsoft Research Podcast.<\/b><\/p>\n Host: Pamela Bhattacharya, welcome to the podcast.<\/b><\/p>\n Pamela Bhattacharya: Thank you. Thank you for having me.<\/p>\n Host: That\u2019s a great name. Does it mean something?<\/b><\/p>\n Pamela Bhattacharya:\u00a0No, I don\u2019t think it has a meaning\u2026<\/p>\n Host: I bet it does somewhere!<\/b><\/p>\n Pamela Bhattacharya: Maybe,\u00a0yeah.\u00a0I don\u2019t know\u2026\u00a0yeah.<\/p>\n Host: Okay. That\u2019s your homework, for the\u2026<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah,\u00a0I think my sister would be better at that. She did this whole family tree for our entire family, so yeah,\u00a0I think she might know.<\/p>\n Host: I\u2019ll get her on the podcast next week.<\/b><\/p>\n Pamela Bhattacharya: Yeah, yeah!<\/p>\n Host: Well, let\u2019s start by getting you situated for our audience. You\u2019re a Principle Applied Scientist at Microsoft. And you\u2019re currently working in the Outlook product group.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Mmm-hmm.<\/p>\n Host: I\u2019ll have you tell us that story a little bit later, but first I want you to tell us, more generally, about what excites you about the work you do and what gets you up in the morning.<\/b><\/p>\n Pamela Bhattacharya: I\u00a0think I\u00a0really believe in Microsoft\u2019s core mission, which is to empower every individual and every organization in the world. And I think productivity is the core to it. And I think, after working in this space for, you know, almost four years now, it\u2019s seems counterintuitive,\u00a0but in terms of productivity, doing less is achieving more.\u00a0Finding that right balance is very important,\u00a0and how we empower people to find their own balance\u00a0\u2013\u00a0because everybody is different, their\u00a0scenarios are different, it\u2019s all contextual\u00a0\u2013\u00a0how we help people find that balance for themselves,\u00a0I think,\u00a0is what excites me and keeps me going.<\/p>\n Host: I want to go in on \u201cdoing less is achieving more.\u201d Can you unpack it a little bit?<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah. It was very personal for me. As I was progressing in my career at Microsoft, I was working on different projects\u00a0and\u00a0even within the same, you know, team I was trying to have impact on different things. And then, you know,\u00a0I had my son and suddenly the life\/work balance, you know, what it meant, changed. But I felt like I still wanted to do impactful work and I felt like the only way I could continue to do impactful work is by choosing problems that are really, really important and doing only one or two of them rather than doing eight or nine of them, which,\u00a0probably,\u00a0I was doing earlier. And I felt like I became more productive by making that conscious decision. Almost every day I have to push myself to think,\u00a0like you know,\u00a0what can I cut? What can I not do and still achieve what I want to achieve?\u00a0And\u00a0I think that\u2019s where we can help build tech for people to make them more conscious and aware of that.<\/p>\n Host: I want to hover at ten thousand feet for a minute and talk<\/b>,<\/b>\u00a0somewhat philosophically<\/b>,<\/b>\u00a0about a pretty big topic, and it\u2019s the future of work.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Mmm-hmm.<\/p>\n Host:\u00a0<\/b>And i<\/b>t was a theme of MSR<\/b>\u2019<\/b>s faculty summit last year and the premise, which I quote from the conference overview is<\/b>,<\/b>\u00a0\u201cNew advances in computing are transforming existing work and productivity paradigms. Tomorrow, we will work in more places, faster, more collaboratively and our output will be evermore thoughtful, creative and impactful.\u201d That\u2019s a big assertion!<\/b><\/p>\n Pamela Bhattacharya:\u00a0Mmm-hmm.<\/p>\n Host: From your perspective, and this is just your perspective, but what\u2019s hope and what\u2019s hype about that statement?<\/b><\/p>\n Pamela Bhattacharya: I think the hope is that if we can all collaborate in a way that makes us individually productive and globally productive\u2026\u00a0Just to give an example, so on days I\u2019m maybe just coding, and I feel productive at the end of the day because I finished my feature, right?\u00a0And on some days,\u00a0I don\u2019t even write a line of code and all I did was brainstorm with other people, you know, help unblock others, you know, white board and try to find solutions,\u00a0and I still feel productive at the end of the day because, you know, I did so much, which cannot be really measured in terms of lines of code or, you know, what I delivered, how many presentations, nothing like that. But I feel that one of the major challenges in the productivity\u00a0space is coming up a very tangible way to measure how productive I was or\u00a0an organization was. You know, we have productive discourse, but at the end of the day,\u00a0how do I feel about it? How does my team feel about it? I feel\u00a0that\u00a0there is a gap in what\u00a0are\u00a0we doing and what we are measuring and how that all ties back as\u00a0a\u00a0feedback\u00a0loop, for example,\u00a0and how we keep improving.<\/p>\n Host<\/b>: I\u2019ve been fascinated by the diversity of research interests among the scientists here<\/b>\u00a0and m<\/b>any of them have self-identified \u2013 I\u2019m not even kidding<\/b>,<\/b>\u00a0they have \u2013 there\u2019s a research grid, where you\u2019re either like Niels Bohr, who was all about basic research<\/b>,<\/b>\u00a0or you<\/b>\u2019<\/b>r<\/b>e<\/b>\u00a0Thomas Edison who was all about applied research<\/b>,<\/b>\u00a0or you<\/b>\u2019re<\/b>\u00a0Louis Pasteur who is some mix of applied and basic. But where do you self-identify on that grid? What kinds of projects fall in your wheelhouse and what value do you think your work brings to the world?<\/b><\/p>\n Pamela Bhattacharya: I think it\u2019s definitely in the applied quadrant.\u00a0I feel\u00a0one thing that I really enjoy is problem solving.\u00a0And\u00a0it could be that I come up with\u00a0the\u00a0most simple solution that works,\u00a0or it could be I worked on it and it\u2019s very complicated and it\u2019s very layered, so on and so forth.\u00a0So for me, the complexity of the solution doesn\u2019t change my approach to the problem. If it works, it works, you know? So that\u2019s what I care most about.\u00a0And we can talk later\u00a0about, you know,\u00a0my journey in\u00a0calendar.help,\u00a0but\u00a0after I joined Microsoft, I moved to a team which was just starting. I was literally the first hire on the team.\u00a0And\u00a0I didn\u2019t think about being in a startup environment or being in that culture or doing something, you know,\u00a0grounds up, but that was a grounds up initiative that we were taking at Microsoft to transform customer support experience and I just loved the experience, you know, having a small team, having much more connection with the leaders and the managers and making sure I really understand what they are trying to achieve and how I am aligned with their goals and so on and so forth. And just the agility of it.\u00a0That\u2019s when I realized that these are the kind of incubation teams that I really find making more impact. I enjoy being in incubation efforts and taking them to completion.<\/p>\n Host: Well, let\u2019s talk about one specific incubation project that you\u2019ve just mentioned. It\u2019s getting a lot of attention right now<\/b>, and it<\/b>\u00a0was called, as you said,\u00a0<\/b>calendar.help<\/b>\u2026<\/b><\/p>\n Pamela Bhattacharya: Yeah.<\/p>\n Host: \u2026but it\u2019s now called Scheduler<\/b>, and y<\/b>ou\u2019ve referred to it as both. I\u2019ll frame it as a personal assistant for people who don\u2019t have \u201cpeople.\u201d Before we get into the specifics on the technology, set up the problem for us. What was the pain point that prompted the project? In other words, why do we need it?<\/b><\/p>\n Pamela Bhattacharya: You might be familiar with Doodle, which is an online scheduling tool.<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya: And\u00a0Doodle did a survey of more than, I believe, fifteen hundred professionals, and what they found is, on average, people in these roles spend more than five hours in scheduling meetings, just back and forth in trying to find the right time. Imagine, five hours, every person, spending every week, to schedule meetings.\u00a0And as you know, even VPs at Microsoft,\u00a0they have their own executive admins that manage their calendars, manage their time, right?<\/p>\n Host: Right. I\u2019ve dealt with them.<\/b><\/p>\n Pamela Bhattacharya:\u00a0So, how do we solve this? And the funny part is\u00a0that, before I joined the project,\u00a0you know,\u00a0I always thought it was scheduling a meeting, but in reality, a meeting has a lifecycle. You initiate a meeting.\u00a0After multiple back and forth, it gets scheduled. And more often than not, it needs to be rescheduled, right?\u00a0Or people need to be added, or, you know\u2026\u00a0you name it. Like you know, it has a life on\u00a0its own. Rarely does a meeting get scheduled, it happens, and it\u2019s done.<\/p>\n Host: I can\u2019t remember the last time<\/b>\u00a0that<\/b>\u00a0that happened.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah. Exactly, even with our podcast, right?<\/p>\n Host: Right\u2026. Snow.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah.\u00a0So, I feel like that\u2019s\u00a0the\u00a0complexity\u00a0it\u00a0brings to it. So, it\u2019s not like once and done.<\/p>\n Host: Sure.<\/b><\/p>\n Pamela Bhattacharya: You know, you don\u2019t know how many times you have to iterate over it.<\/p>\n Host: Okay. So, everything you\u2019ve just said, I bet, is going to resonate with everyone in our audience, except the people who have people. And so that pain point is real. I\u2019m just nodding my head, no one can see it, but I\u2019m going yup, absolutely. So, tell us about your solution, which is now called Scheduler. It\u2019s an intelligent personal digital assistant, IPDA, for those TLA fans, and it brings recent advances in machine learning\u00a0<\/b>\u2013\u00a0<\/b>and that\u2019s the intelligence<\/b>\u00a0\u2013\u00a0<\/b>to the field of PDAs, personal digital assistants. So I\u2019m going to let you run with this because you do a fantastic job of explaining it, but don\u2019t be afraid to get technical.<\/b><\/p>\n Pamela Bhattacharya:\u00a0OK, so let me give a little bit of background. So, Scheduler started with the idea of building a gig economic platform for executive assistants,\u00a0and\u00a0that\u2019s where it started in FUSE labs, and soon they realized that,\u00a0of all jobs that executive admins need to do, scheduling is the pain point,\u00a0like we discussed earlier. So essentially,\u00a0Scheduler started\u00a0with\u00a0being this human and AI hybrid platform to enable scheduling much more easier. So, as a user I can off-load it to the intelligent assistant and they take care of it. So, the user experience is really\u00a0very much like how,\u00a0say,\u00a0you see our\u00a0VPs interact. So, say I\u2019m having an email conversation with my VP Gaurav and, you know,\u00a0at some point he might say hey, okay,\u00a0let\u2019s just meet and chat, you know.\u00a0It\u2019s getting complicated.\u00a0And then he\u2019ll\u00a0cc\u00a0his admin, Crystal, and\u00a0he\u2019ll\u00a0say Crystal, can you find us a time?\u00a0And he might give more constraints. Find us a time next week,\u00a0or find us a time and add somebody else who is not on the thread, right?\u00a0And then\u00a0on, what\u00a0Crystal\u00a0does, Crystal\u2019s job is to understand those constraints and schedule the meeting.\u00a0Now let me get to the technical part. So, the assistant or the service gets this email, but this email contains a lot of information that\u2019s not relevant to scheduling, right?\u00a0You know, say we talked about the project,\u00a0we talked about\u00a0deadlines,\u00a0we talked about\u00a0other meetings, but to the assistant,\u00a0you are very specifically saying about scheduling one meeting. So, the first technical problem that we solved was, given a document, how do you find out what are the relevant sentences for a task at hand? So here the task\u00a0at hand\u00a0is\u00a0the scheduling, so this was a ranking, you know, algorithm that we came up with in finding relevant sentences for that. Okay, so that\u2019s step one. Step two, once you have found the relevant sentences, then it\u2019s about understanding intent. What are you trying to do? Oh, he is\u00a0trying to schedule one meeting, recurring meeting, an online meeting, a phone call, a lunch\u2026 because all those will define how you\u00a0choose\u00a0the times or what you do with it, right? And it\u2019s very contextual. So, you might say let\u2019s have a lunch meeting, right? So,\u00a0then\u00a0the assistant needs to understand that okay, I can only look between times,\u00a0say,\u00a0between eleven thirty and one because nobody has lunch either at ten or at four, right?\u00a0But\u2026\u00a0So, understanding the context, understanding what are the constraints that apply to this meeting, you know, your intent,\u00a0is the second step. And the third step is now,\u00a0what do I do with this, right? How do I process it in a way and what is my final outcome? So, for example, one of the outcomes could be, ok,\u00a0I\u2026\u00a0say a recruiter is trying to schedule a meeting with a candidate, but the assistant doesn\u2019t have access to the candidate\u2019s calendar, right?<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya:\u00a0So, they have to reach out, just like an admin would reach out and say hey, Pamela is available times X, Y, Z, you know, does any of this work? And then reply again in\u00a0absolute natural language saying either yes, all that works, or no, it doesn\u2019t work, how about something else? Or they might say, I can do the first and the third, right?\u00a0Or,\u00a0you know,\u00a0even more complicated,\u00a0it becomes like\u00a0if,\u00a0say,\u00a0we give three options, two on Tuesday and one on Thursday, they just say Tuesday works. So now, we have to do the job of mapping Tuesday with the options and, you know, so it\u2019s a whole other complicated system that\u2019s out there. And then again, you know, as I said earlier, meeting has a lifecycle, so it might get rescheduled, you might need to cancel it, you might need to add more people, you might need to drop people because not everybody is available, there\u2019s no mutual availability, so on and so forth.<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya:\u00a0So, that\u2019s one layer of natural language understanding, you know, parsing, interpreting. And then there\u2019s another layer of preferences, right?\u00a0Personalizing the assistant for you because if I,\u00a0say,\u00a0hypothetically,\u00a0had a human admin,\u00a0they would know what times I prefer.<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya:\u00a0I might not be a morning person at all.<\/p>\n Host: And they know that.<\/b><\/p>\n Pamela Bhattacharya: They would know that,\u00a0you know,\u00a0over time. But if I have a meeting with somebody, say,\u00a0in MSR India, they wouldn\u2019t hesitate to schedule a meeting at eight a.m. because,\u00a0you know,\u00a0that\u2019s a more human time to have a meeting, right?<\/p>\n Host: Right. For everyone involved.<\/b><\/p>\n Pamela Bhattacharya: Exactly. So, again, how do you have personalization? But how do you add context to it?<\/p>\n Host: Okay.<\/b><\/p>\n Pamela Bhattacharya: So, that, you know, it\u2019s the best of all worlds.<\/p>\n (music plays)<\/i><\/b><\/p>\n Host:<\/b>\u00a0<\/b>B<\/b>efore we get into the open problems which you\u2019ve alluded to<\/b>,<\/b>\u00a0which involve personalization, disambiguation, entity extraction, intent, negation, all of that stuff<\/b>\u00a0\u2013 a<\/b>nd I want to talk\u00a0<\/b>to you\u00a0<\/b>about that<\/b>\u00a0\u2013 f<\/b>irst tell us what this product does.\u00a0<\/b>How does it work?\u00a0<\/b><\/p>\n Pamela Bhattacharya: It\u2019s very seamless.\u00a0It\u2019s essentially a similar experience to having a human admin where you just\u00a0cc\u00a0the admin, which, in this case, is an email address,\u00a0cortana@calendar.help,\u00a0and the service takes over from there. And anything, any\u00a0new instructions you need to add, you just send an email to your assistant.<\/p>\n Host: Okay, so the assistant is actually an intelligent agent\u2026<\/b><\/p>\n Pamela Bhattacharya: Yup. Yup.<\/p>\n Host: \u2026and the intelligent agent is going to try to do all\u00a0<\/b>of\u00a0<\/b>the things that a human would.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah, yeah.<\/p>\n Host: Let\u2019s talk about the problems that are still out there\u00a0<\/b>that\u00a0<\/b>you\u2019re working on.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah,\u00a0so it\u2019s a\u2026\u00a0we are still working it, right?\u00a0And it\u2019s a mixture. So initially,\u00a0we look at your calendar and we try to understand what are your meeting behaviors and preferences, right? But then, those all change over time. And again, those have, you know,\u00a0layers of context with it, right?\u00a0So I might have morning meetings twice a week but those are only for meeting with people in India, with the MSR team there. Or I might have late meetings,\u00a0and that might be only for meetings with the\u00a0Suzhou\u00a0team, right?\u00a0And then during the day, I might have one-on-ones clustered, but say other technical meetings might have more space between them because I might need to come from one meeting, you know, just take some time out for myself and then have a next meeting, so how to balance all these things out. And that\u2019s where we are, you know,\u00a0trying out reinforcement learning to personalize these recommendations based on the context, you know, so that\u2019s one area\u00a0that\u00a0we are investing in.<\/p>\n Host: One of the most interesting areas that you talked to me about was negation\u2026<\/b><\/p>\n Pamela Bhattacharya:\u00a0Mmm-hmm.<\/p>\n Host: \u2026and how a machine deals with what a human can understand really easily. Why don\u2019t you explain that?<\/b><\/p>\n Pamela Bhattacharya: Yeah, so, the other complicated part of the entity extraction that I talked to you about earlier is how the same semantics can be expressed in so many different ways. So for example, I might say, hey,\u00a0I can meet you Gretchen next week, except Thursday. Right? That results to exactly the same time range,\u00a0if you will, versus if I had said I can meet you next week. I am OOF on Thursday. So that\u2019s a very implicit way of saying I\u2019m not available on Thursday. So, in both these cases, I can meet next week except Thursday, and,\u00a0you know,\u00a0breaking it into two sentences where you have an implicit way of saying you\u2019re unavailable, the concept of negation is something, you know, that is very important for us because those are the times you do not want the meeting to happen.<\/p>\n Host: Right.\u00a0<\/b><\/p>\n Pamela Bhattacharya: Yeah, and that\u2019s something, you know, again, we are collaborating with MSR and we also have our home-grown solutions that we are trying to do better at.<\/p>\n Host: Right, right, right. All right. We\u2019re talking about people<\/b>, right?<\/b>\u00a0<\/b>H<\/b>ow do we get bodies in a room?\u00a0<\/b>And if<\/b>\u00a0it\u2019s a phone meeting that\u2019s one thing. But let\u2019s talk about other resources for a second because it isn\u2019t just people, it\u2019s how many people, it\u2019s whether the rooms are available\u2026\u00a0<\/b>for example,\u00a0<\/b>there might be people from other buildings that need to come.<\/b>\u00a0So what about these other resources?\u00a0<\/b>How are you dealing with that?<\/b><\/p>\n Pamela Bhattacharya: Yeah. You bring up a\u00a0very\u00a0good point about location intelligence, right? So, resource booking is such an important part of meeting scheduling. Imagine having a meeting time, but then not a room, right?\u00a0I mean, how often do we change the meeting time because a conference room is not available?<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya: Right? So, that\u2019s\u00a0really\u00a0crucial. So we also support conference room booking,\u00a0resource booking. So, you know, as a user, you can just say\u00a0that, hey, book a room for me, or book a room in Building 32 or in Building 99, right? And the agent will then extract the intent that you need a room, and then the entities associated with that intent, that oh,\u00a0it\u2019s supposed to happen in Building 99. Or if you don\u2019t say something, we just go with your, you know, most recently used or preferred rooms.<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya: You also can\u00a0specify to the agent a list of your preferred rooms. Say our VPs mostly they\u2019ll have their own preferred rooms, you know, reserved rooms.<\/p>\n Host:\u00a0<\/b>Mmm<\/b>-hmm<\/b>.<\/b><\/p>\n Pamela Bhattacharya: So, you can also do that version with the agent. And then we try to book you a room. And if you are not able to find a room, you know, we just tell you we were not able to, but internally what we try to do is, we try to find a time when a room is available.<\/p>\n Host: Interesting.<\/b><\/p>\n Pamela Bhattacharya:\u00a0So, the ranking algorithm changes because now it has a different constraint to it, which that it requires a room.<\/p>\n Host: I was just going to ask you about constraints because each layer is a new constraint and I would hope, as a person I would be, yeah, I get that and I know they are all here in\u00a0<\/b>Building 99, blah, blah,<\/b>\u00a0blah,<\/b>\u00a0but there is somebody from over in 32 that needs to come over here\u2026<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah, yeah, to add to that, travel time\u2026<\/p>\n Host: Right!<\/b><\/p>\n Pamela Bhattacharya:\u00a0Travel time is something that we haven\u2019t started supporting yet, but that\u2019s, you know, one of our biggest things that we want to invest in and you know we want to solve, because the agent should be intelligent enough to say that, oh, all they are free from eleven to twelve. You have a meeting until eleven in Building 99. You cannot make it at eleven to Building 32, right?\u00a0So\u00a0we need to give that travel time for meetings.\u00a0So, today, we don\u2019t have that intelligence in our system.<\/p>\n Host: Well, let\u2019s talk a bit about where you are with this because that leads in really well to my question about automation. And you have a wonderful explanation for levels of automation and where Scheduler lives now\u2026<\/b><\/p>\n Pamela Bhattacharya: Yeah, yeah.<\/p>\n Host: \u2026and where you\u2019re heading by tackling the problems we\u2019re just talking about.<\/b><\/p>\n Pamela Bhattacharya: Yeah, yeah!\u00a0So, let me first, you know,\u00a0for our audience,\u00a0give a brief about the levels of automation so that everybody is on the same page. So, we draw the analogy from automation in cars, right? So, in the automobile industry,\u00a0the Society of Automobile Engineers, they have done a great job in creating these levels, right? Let me go through the levels. Level Zero is like, you know, you are in complete control, there is no intelligence,\u00a0you know,\u00a0no assistance, nothing. Level One is like most smart cars. So, you are still in complete control, but you have some assistance, right?\u00a0Level Two, Level\u00a0Three is more. The assistance is increasing.\u00a0You can get, you know,\u00a0sensors when you have a blind spot or your wipers turn on automatically when it\u2019s raining\u2026 Again, you are doing less to do more.\u00a0And then Level Four, you are slowly giving away, delegating more and more of the task of driving to the agent, but you are still in control. You can respond if there is an emergency. And Level Five, you are just in your back seat and sleeping.<\/p>\n Host: Autopilot.\u00a0<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah. Autopilot. There you go. So in terms of your personal agents, that\u2019s where the spectrum is, right?<\/p>\n Host: Okay.<\/b><\/p>\n Pamela Bhattacharya: And we are probably at Level Two right now, right?\u00a0So\u00a0I think this framework is applicable to more than\u00a0Scheduler or anything intelligent because I feel it\u2019s a great way to set user expectations.<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya:\u00a0So imagine if I went to buy an intelligent car and I didn\u2019t know where in the spectrum it falls,\u00a0and I bought a Level Two car expecting\u00a0it\u2019s\u00a0a Level Five car, I\u2019ll be disappointed, right?<\/p>\n Host: Or angry\u2026<\/b><\/p>\n Pamela Bhattacharya: But if I knew\u00a0that\u00a0I\u2019m buying a Level Two car,\u00a0I know what I can expect.<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya: Similarly, with digital assistants. So, many of our users, you know, in the initial days,\u00a0they would expect Scheduler to do things which, it\u2019s probably a Level Five expectation. So for example, can you reschedule a different meeting on my calendar to accommodate this meeting? We don\u2019t support that, right? But they assume as\u2026 because humans can do that, human admins can do that. When you share your calendar with them, when you give them access, they can go to your calendar and reschedule meetings, right? That\u2019s a Level Five. We don\u2019t support that. So we did sense frustration from people because they were not aligned with what the technology was,\u00a0what\u00a0the product was.<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya: So this gives us a great way to align our product,\u00a0you know,\u00a0and where we are with the user expectations, because a lot of happiness in using something or doing something comes from having the right expectations.<\/p>\n Host: I\u2019m loving your phrase, \u201cWe don\u2019t support that.\u201d I think I\u2019m going to start using it instead of, I can\u2019t do that. I don\u2019t support that\u2026<\/b>!<\/b><\/p>\n (music plays)<\/i><\/b><\/p>\n Host:\u00a0<\/b>Scheduler is literally a poster child for tech transfer stories. Give us a short biography of the project and your association with it. Where was it born<\/b>, w<\/b>here does it live now, and where is it in the pipeline to shipping so I can use it?<\/b><\/p>\n Pamela Bhattacharya: Yeah, so Scheduler,\u00a0back then,\u00a0started as\u00a0calendar.help. It started in the Future of User Social Experiences\u00a0Lab, also known as the FUSE Labs at\u00a0Microsoft Research.\u00a0As I said earlier, it started as a, you know,\u00a0gig economic platform to help executive admins, but then\u00a0the original team quickly realized that, you know,\u00a0scheduling is a big opportunity in that domain,\u00a0and that\u2019s where it was born. I joined the team in October 2016 and that was when Microsoft acquired Genie, which was a startup in the Bay area\u00a0and this was founded by my current manager, Charles,\u00a0and his co-founder Ben. And Microsoft acquired Genie and the goal was that we will transfer this over to Microsoft Outlook and it will become a product there. I didn\u2019t know where the project was going.\u00a0You know,\u00a0I heard about the announcement and\u00a0I\u00a0read about Genie and\u00a0I read\u00a0about\u00a0calendar.help.\u00a0I didn\u2019t know about these projects or these tools or anything.\u00a0And\u00a0I just wanted to be a part of it. Yeah.<\/p>\n Host: So where is it now, in the pipeline?<\/b><\/p>\n Pamela Bhattacharya: So, yeah, so\u00a0we have been in a public preview program since November of 2016. And last Ignite, November 2019, we announced we are going GA sometime this year.<\/p>\n Host: Well, the core technology behind Scheduler seems like it could have other applications. You\u2019ve actually kind of alluded to that already. How could you envision using the core technology that\u2019s behind this beyond calendars and scheduling?<\/b><\/p>\n Pamela Bhattacharya: Yeah, I think the core technology is about understanding intent in natural language. And today we only support emails as a written form of natural language, but it can be spoken, right? And also, understanding context and preferences. So, if I had to layer these, right, the first would be understanding intent. Second is,\u00a0how\u00a0do you extract the right information knowing the intent of a user? And then how do you find out what their preferences would be in that context and then get the job done?<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Right? And that can be applied to so many things. Reserving restaurants, finding the right times to meet your family,\u00a0or,\u00a0you know,\u00a0when everyone is available, you know\u2026\u00a0there are a myriad of scenarios where you can use that.<\/p>\n Host: Well, we\u2019ve talked about what gets you up in the morning, but this is the part\u00a0<\/b>of the podcast\u00a0<\/b>where I ask what keeps you up at night. So, I think about the work you\u2019re doing, helping the assistant-less among us get more done with their time. Even if we\u2019re doing less by achieving more! And that all feels good to me, but I\u2019m sure there are some consequences, intended or otherwise, that might need to be addressed. So, what kinds of things keep you up at night<\/b>, a<\/b>nd what are you doing about it?<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah. So, there are really two areas that makes me anxious about,\u00a0you know,\u00a0the progress we are making and what we can do better. So one is definitely, you know, privacy, user privacy, and ethics and what I think is currently termed as Compliance Constrained AI, right? You have to put the user\u2019s privacy as a first class citizen in everything you are doing. So I think that\u2019s definitely a big area where I know many, many, many teams in Microsoft is\u00a0investing, and I think we have a long way to go there, you know. How can we build even powerful models but making sure that we are honoring the privacy of the users?<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya: And I think a second bucket is more about user education and how can we bring something intelligent, but so seamless for people to use who might not know anything about tech, right? There are all these complex pieces that we are building together, but how do you make it very easy for people to start using it?\u00a0And\u00a0I\u2019ll give the context of Scheduler itself, right? Many times, when,\u00a0say,\u00a0the organizer of the meeting, you know, they have cc-ed Cortana and Cortana had reached out to the invitee saying that hey,\u00a0you know,\u00a0something,\u00a0I need this information from you. Many a times, those invite people who receive that invite from Cortana do not know how to react to it. Should I write an email, it\u2019s an agent, it\u2019s a\u00a0bot, like you know,\u00a0what should I do with it, right?<\/p>\n Host: Right.<\/b><\/p>\n Pamela Bhattacharya: So, it doesn\u2019t scale for us to go one-on-one and tell everybody how you use the product, right?\u00a0So, how can we make that so seamless?\u00a0Because,\u00a0in the core,\u00a0the service itself is so complicated, right, but then how do you\u00a0just\u00a0almost like flip the coin and the person who is using it, or is on the other side of\u2026\u00a0not just the user, but,\u00a0you know,\u00a0the invitees of the meeting or anybody involved, you know, in the service, to them it feels seamless. I feel that is very core and today I feel there are a lot of missing parts where we could do more.<\/p>\n Host: Well, it\u2019s story time, Pamela, and I would love to hear yours<\/b>!<\/b>\u00a0Tell us a little bit about yourself. What got you started along the path to automating my scheduling problems and how did you end up at Microsoft?<\/b><\/p>\n Pamela Bhattacharya:\u00a0So\u00a0I came to the US to get my PhD and along the way I did an internship at Microsoft and I got a full-time offer. It was more a personal decision, at that point,\u00a0to join Microsoft,\u00a0because me and my husband,\u00a0uh you know, my\u00a0boyfriend at the time, we were in a long distance relationship for almost six years, and he got an offer from MSR and we felt like,\u00a0you know,\u00a0this was a great opportunity for us to be together. And the reason I say that,\u00a0is there was no applied science\u00a0role\u00a0eight years back at Microsoft. You know, data, machine learning, AI\u2026\u00a0these were not,\u00a0you know,\u00a0first class citizens that,\u00a0you know,\u00a0as a company, we were looking at or investing in. So I didn\u2019t really know how\u00a0my\u00a0muscle\u00a0that\u00a0I built at grad school during my PhD would be useful here, but luckily for me,\u00a0you know,\u00a0in almost\u00a0like\u00a0one and a half years after joining Microsoft,\u00a0we had this applied science role. And that\u2019s,\u00a0you know,\u00a0how I started with this team I was talking about earlier as the first\u00a0person, the first\u00a0hire\u00a0on the team\u00a0where we wanted to use machine learning and AI to transform customer interactions and customer support in Office 365.<\/p>\n Host: Okay.<\/b><\/p>\n Pamela Bhattacharya: So I believe it just worked out for me in that way and\u00a0like\u00a0how I always wanted to be an applied researcher.\u00a0So that was about how I came to Microsoft and found a role that I really enjoy doing on a day-to-day basis. And then,\u00a0you know,\u00a0about Scheduler, I saw the announcement over email, you know,\u00a0when Microsoft has an acquisition, they usually send out an email to the larger org that hey, we are doing this, we are investing in this. And I found out and,\u00a0you know,\u00a0I was very lucky that my VP,\u00a0Gaurav, who is the VP for Outlook, he introduced me to Charles, who is my current manager.\u00a0And you know, because I was so interested in working in this space,\u00a0and things just happened from there. I mean,\u00a0yeah.<\/p>\n Host:\u00a0<\/b>So, w<\/b>here did you do your doctorate?<\/b><\/p>\n Pamela Bhattacharya:\u00a0From\u00a0University of California, Riverside.<\/p>\n Host: So you stared in India, undergrad and everything. Where in India were you living?<\/b><\/p>\n Pamela Bhattacharya: So I was born and brought up in the suburbs of Calcutta. And\u00a0my, yeah,\u00a0my family now moved to Calcutta, but yeah I was born there. I\u00a0think, again,\u00a0it ties so much back to how much my family believed in STEM education. And my parents, you know, from a very early age,\u00a0they\u00a0made sure that, you know,\u00a0education is important. Math is important. And actually my dad,\u00a0I believe,\u00a0was more interested for me to pursue computer science. Both my parents, actually, then. I don\u2019t think I knew what it meant, you know, when I signed up for coding classes.<\/p>\n Host: Sure.\u00a0<\/b><\/p>\n Pamela Bhattacharya:\u00a0And, but I enjoyed it, you know, and\u00a0I think\u00a0that\u2019s how things went.<\/p>\n Host: What\u2019s one thing people might not know about you, whether it\u2019s a personal characteristic or a\u00a0<\/b>life-<\/b>defining moment or a personality trait or a side quest that may have influenced your decision to become an applied scientist.<\/b><\/p>\n Pamela Bhattacharya:\u00a0So I think\u00a0I did not start thinking I\u2019ll become an applied scientist, right? What I did know is that I had a lot of role models, right?\u00a0Like,\u00a0people I wanted to be like.\u00a0And\u00a0I wanted to make an impact the way they were making, right? And\u00a0I think that all these people that I really admire, it took\u00a0them\u00a0a lot of hard work to get there. What we often think\u00a0of, like,\u00a0somebody is naturally good at it, but they have put in a lot of effort to get there.<\/p>\n Host:\u00a0<\/b>Mmm<\/b>-hmm.<\/b><\/p>\n Pamela Bhattacharya:\u00a0So that was one thing. And the second thing is that,\u00a0if I role model someone, I don\u2019t have to be exactly what they are doing. I have to,\u00a0kind of,\u00a0extract out what,\u00a0in them,\u00a0do I really appreciate and how can I build that muscle?\u00a0So I felt\u00a0like\u00a0it was more about\u00a0what are they doing\u00a0and\u00a0how are they doing and how can I apply that to my scenario and take it from there?\u00a0So I think that\u00a0\u2013\u00a0internalizing that\u00a0\u2013\u00a0helped me in my path\u00a0forward.<\/p>\n Host: It sounds like you\u2019re pretty analytical about things.<\/b><\/p>\n Pamela Bhattacharya: It\u2019s funny you say that because my manager told me the same thing in my last Connect.\u00a0Because we were talking about a very different problem,\u00a0like\u00a0not even with the product, and he said,\u00a0you know,\u00a0Pamela, you are really analytical. Yeah, I guess I am, yeah!<\/p>\n Host: As we close,\u00a0<\/b>and I\u2019m<\/b>\u00a0sad\u00a0<\/b>to close\u00a0<\/b>because it\u2019s so delightful talking to you<\/b>\u2026<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah, same here, thank you!<\/p>\n Host:\u00a0<\/b>\u2026<\/b>I want to give you the opportunity to talk to your grad school self, the one who didn\u2019t know where her career path would take her and would love some advice from her future self. So what would you say to you<\/b>,<\/b>\u00a0and by extension to those in our audience who are where you were then right now?<\/b><\/p>\n Pamela Bhattacharya: I think perseverance is really important,\u00a0and sometimes we don\u2019t realize even people who we know as really successful, how much failures they have gone through.\u00a0And\u00a0I feel that is something\u00a0that I\u00a0would tell my younger self is, just coming back every day and trying to set the right expectations from yourself, being\u00a0kind to yourself and whatever your dreams are, you should go for them and you should just keep doing it.<\/p>\n Host: And not give up.<\/b><\/p>\n Pamela Bhattacharya:\u00a0Yeah. Not give up.<\/p>\n Host:\u00a0<\/b>It reminds me of a phrase I\u2019ve heard that I really like<\/b>.<\/b>\u00a0\u201cSuccess isn\u2019t permanent. Failure isn\u2019t terminal.\u201d<\/b><\/p>\n Pamela Bhattacharya: Absolutely. Yeah, yeah. That\u2019s great. That kind of summarizes.<\/p>\n Host: Pamela Bhattacharya, thank you so much for joining us today. It\u2019s been an absolute delight.<\/b><\/p>\n Pamela Bhattacharya: Thank you, it\u2019s an honor. Thank you for having me on your show.<\/p>\n (music plays)<\/i><\/b><\/p>\nRelated:<\/h3>\n
\n
\nTranscript<\/h3>\n
![\"photo](\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/02\/Research_Podcast_Bhattacharya_Site_1400x788-960x540.png\")
","byline":"","formattedDate":"February 12, 2020","formattedExcerpt":"In a world where productivity is paramount and only a handful of people have personal assistants, many of us are frustrated by the amount of time we spend in meetings, and worse, the amount time we spend planning, scheduling and rescheduling those meetings! Fortunately, Dr.…","locale":{"slug":"en_us","name":"English","native":"","english":"English"},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/636450"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/users\/39507"}],"replies":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/comments?post=636450"}],"version-history":[{"count":7,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/636450\/revisions"}],"predecessor-version":[{"id":886941,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/636450\/revisions\/886941"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/636462"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=636450"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/categories?post=636450"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/tags?post=636450"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=636450"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=636450"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=636450"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=636450"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=636450"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=636450"},{"taxonomy":"msr-promo-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-promo-type?post=636450"},{"taxonomy":"msr-podcast-series","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-podcast-series?post=636450"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}