Bot on Thomy Gölles

SelectedTech Webinar 3 - CI/CD with your bot and Azure DevOps

Wed, 17 Jul 2019 21:34:35 +0000

The SelectedTech Webinar 003 about using CI/CD with your bot and Azure DevOps is now available. We recorded the session on July 4th 2019. Check it out and leave a comment or send a tweet or message if you have any questions.

In this episode Stephan demos how to create a build and release pipeline within Azure DevOps. We discuss the two different configurations for build and release and talk about how to reuse your config for different target systems. The goal is to switch your target environment with Azure DevOps and to be able to deploy your bot to different Azure tenants just by clicking through the release pipeline.

SelectedTech Webinar 4 - QnA Maker Postman Collection

Wed, 17 Jul 2019 14:27:35 +0000

The SelectedTech Webinar 004 is about how to use Postman to talk to the QnA maker API. Postman streamlines the development process and captures a single source of truth about your APIs while also simplifying collaboration across your teams and organisations.

QnA Maker is a cloud-based API service that creates a conversational, question and answer layer over your data. QnA Maker enables you to create a knowledge-base(KB) from your semi-structured content such as Frequently Asked Question (FAQ) URLs, product manuals, support documents and custom questions and answers. The QnA Maker service answers your users’ natural language questions by matching it with the best possible answer from the QnAs in your Knowledge base.

SelectedTech Webinar 2 - QnA Bot Local Development

Mon, 08 Jul 2019 17:22:22 +0000

The SelectedTech Webinar 002 about local QnA bot development and deployment to Microsoft Azure and Microsoft Teams is now available. We recorded the session on July 4th 2019. Check it out and leave a comment or send a tweet or message if you have any questions.

In this episode we take the QnA bot from webinar 001 and download the code to a local machine. After showing how to run the bot in a local IIS we use ngrok for tunneling access for a Teams deployment. After making a minor code change we republish the bot and make sure everything is still working. The episode ends with Rick asking Stephan about using Azure DevOps for CI/CD which will be our topic for Webinar 003.

SelectedTech Webinar 1 - A No Code QnA Bot In Microsoft Teams

Mon, 20 May 2019 19:43:30 +0000

Welcome to our first SelectedTech Webinar. We are four MVPs from Europe and we decided to start our own webinar series. What we want to show is how we do stuff. We meet online and hit the record button. There is not a lot of time for preparation or planning for our webinars. As a result, you will see mistakes and errors, not high polished content. We strongly believe that showing how to overcome those hiccups adds additional value to the content.

SharePoint Saturday Barcelona 2018 + "El Clasico" Demo

Sun, 28 Oct 2018 16:52:59 +0000

Yesterday I had the pleasure to present at SharePoint Saturday Barcelona. The whole of 2018 we were kind of touring with our “lunch bell” session. It all began at SPS Barcelona 2017 where a part of the session on PnP was a demo of the lunch bell. The feedback was really great and the idea of creating a session around this topic alone was borne. But all good ends at some point in time in our case this was one of the last times we actually present this topic. We have a full backlog of ideas for 2019 and after ESPC we won’t present this anymore. There is also a real-world reason behind the decision. We as Solvion move to a new office by the 1st of January, so there isn’t a need for this kind of lunch bell anymore. But let’s see what the new office provides and what ideas will be there after moving. Here the slides from yesterday: Back to SharePoint Saturday Barcelona. There is also another big event currently in Barcelona, El Clasico. The football match between FC Barcelona and Real Madrid. Last week in Brussels Edin asked me to use this match for one of my demos. It took me a few hours yesterday, but then I was able to showcase a SharePoint Framework ListView Command Set extension that takes the selected image in a document library and askes Azure Computer Vision for a description of the image. We tried the Command Set with a few images from footballers and got back a friendly message when the Command Set detected a local player from Barcelona. On the other hand, if it detects a certain player from Real Madrid, we just update the image with a logo from Barcelona just to make sure the audience in Barcelona isn’t starting throwing tomatoes at me presenting ;) I will put the complete project code on Github next week, but here the important part with the actual “logic” behind:``` @override public onExecute(event: IListViewCommandSetExecuteEventParameters): void { switch (event.itemId) { case ‘COMMAND_1’: { console.log(“COMMAND_1 clicked”); const itemId: number = event.selectedRows[0].getValueByName(‘ID’); var serverRelativeUrl = “”;

HappitChat: The Role of AI in the Modern Enterprise

Wed, 12 Sep 2018 12:52:25 +0000

At this#HappitChat we will discuss the role of AI in the modern enterprise with our guests from Solvion. The recording will take about 30 minute and our experts will share with you:

How can we incorporate AI into our daily work and collaboration?
How can AI support employees’ work today?
Examples of AI in the modern workplace.
Host: Jussi ‘Juicy’ Mori, Happit
Guests: Thomy Gölles, Solvion | Stephan Bisser, Solvion

SharePoint Saturday Warsaw 2018

Tue, 19 Jun 2018 07:50:04 +0000

Last weekend Stephan (cloudguy.pro) and I had the pleasure to present at SharePoint Saturday Warsaw. The local crew around Edyta managed to get one of the best locations ever. The whole event took place in the local football stadium of Legia Warsaw. Our session was in their press conference room where on a regular day the coach or the players answer questions from the press. [ngg_images source=“galleries” container_ids=“2” display_type=“photocrati-nextgen_basic_thumbnails” override_thumbnail_settings=“0” thumbnail_width=“240” thumbnail_height=“160” thumbnail_crop=“1” images_per_page=“6” number_of_columns=“3” ajax_pagination=“1” show_all_in_lightbox=“0” use_imagebrowser_effect=“0” show_slideshow_link=“1” slideshow_link_text="[Show slideshow]" order_by=“sortorder” order_direction=“ASC” returns=“included” maximum_entity_count=“500”] In our session, we once again showcased how AI in the modern workplace can help in creating an engaging environment for users. In addition to the end user value, we also focus on how a creative idea can be used to learn new things in the fields of Office365 or Azure in a natural way. Our use case of the food truck monitor probably will never be sold to a customer, but you can imagine how much we learned about the technology included during the setup of the demo. Since a few weeks, we try to record all our sessions on video (Go check out our Youtube channel). If you are interested, start watching, for those with a smaller time budget the slides are also available here.

SharePoint Saturday Zurich 2018

Sun, 03 Jun 2018 16:52:05 +0000

Last Saturday we (Stephan Bisser and myself) had the pleasure to be part of SharePoint Saturday Zurich. Jussi Mori, Samuel Zürcher and Stefano Tempesta did a great job in organizing once more an epic event packed with the latest and greatest on SharePoint, Office365 and the relating Azure areas. In his keynote, Vesa Juvonen from Microsoft took us on a brief journey through the history of SharePoint. The screenshots from the portal landing pages of version 2007 and 2010 took some of us in the audience way back in time. Somehow later in the day or to be more precise in the evening, some of the speakers realized how long they have been around SharePoint and what conclusion can be drawn out from this fact. Sorry Maarten & Vesa ;) Stephan and I showed our 21st-century lunch bell with a live demo in Zurich. The lunch bell recognizes a special food truck daily at our driveway and notifies our users in Microsoft Teams or Skype for Business. As we use Cognitive Services for our system with pictures of our office back home it’s actually not that easy to demo the system on the go. To still be able to show the features of Azure we decided to make a mobile version. This means during the session we are trying to detect this little food truck: In terms of preparing the demo, we need to take sample pictures in the actual room of the session to train the model. That means instead of lunch break Stephan and I took the opportunity to take around 60 images of the table on the main stage to later on demo the system within the environment. This type of demo is always special and time intensive to prepare, but when the deadline gets closer and closer the fun starts, right? Thanks again to the super hero organizer of the event, you did an awesome job and we are already looking forward to next year!

Global Azure Bootcamp Switzerland 2018

Sat, 21 Apr 2018 14:13:35 +0000

Today I had the pleasure of co-presenting with my Colleague Stephan at the Global Azure Bootcamp Switzerland in Zürich. As there has been a recent addition to Stephan’s family he wasn’t able to come to Zürich this time, but in today’s world you just take a tiny Jabra USB speaker on the go and ready is the remote presenting interface in Skype for Business. This post is a little follow up on the session, sharing some background links. Also, there were some questions afterwards and instead of writing different emails in private, I decided to blog about it and just share the link. Bringing the spirit of “sharing is caring” from SharePoint / Office Dev Patterns and Practices to the Azure community, right? The session was about Stephan’s project called “Homie” – A language understanding bot for smart homes. We talked about how you can use a chatbot based on the Microsoft Bot Framework and LUIS and wire it up to use AzureIoT Hub for communication with a RaspberryPi. The idea is that you tell the chatbot to turn the lights on or off in a given room. The bot uses LUIS to identify your intent, recognises commands and objects and triggers a command flow in AzureIoT Hub. From there it’s only a small step to actually get the command to the RaspberryPi and switch on the lights on our demo led module. The whole setup looked something like this: This was our test run before the session. And here the working system: Everything turned out good, the demo gods were kind and even the remote setup was flawless. As there were some questions afterwards, I just do a little list of links here pointing to different articles for further reading: A comparison of the different cognitive services out there Tech Showcase: Custom Vision Service Computer vision made easy: From pre-trained models to Custom Vision, Microsoft Cognitive Services ha Custom Vision from Cognitive Services: easily build a custom image classifier Building image classification using the Microsoft AI platform - BRK3334 <– The one with the background info on machine learning Sonya Koptyev’s tweet on what demos to expect from Build on AI (drones!!!11!!) A complete guide on the solution on Stephans’ blog and a German version here Azure Sphere announcement from April 16th, 2018 Jeremy Thake’s post on the chatbots and how they don’t replace apps And of course our slides: Big thanks and kudos to the Azure Zürich User Group for setting this up and let’s hope for more events in the future.

Theodore - a bot for creating a team in Microsoft Teams

Fri, 17 Nov 2017 23:07:20 +0000

My last posts were all about getting up to speed with the Microsoft Teams PowerShell cmdlets released two weeks ago. Running those cmdlets in an Azure Function creates some nice options in terms of creating a team in Teams (still, the wording is just awkward). I’m sure you read about all the governance discussion that services like Teams or Groups start in each environment. Only with proper tooling, you can manage the process of orchestrating your different collaboration features in Office365. To take this idea to the next level, what about using a bot in Teams to handling a users request of creating a new team. This post describes how you can create a bot for your users that will ask for some data in order to create a new team in Microsoft Teams. The bot will call an http triggered Azure Function that is using PowerShell to handle the requests and has the new Teams cmdlets deployed to be able to create a new team in our tenant. To code in the git hub repo is based on an example project (https://github.com/Microsoft/BotBuilder-Samples) of a dot.net bot that shows how the different dialogs in a bot can be wired together. The solution is quite simple and of course no way production ready: An example dialog of the bot looks like this: To create this kind of dialog you simply start the interaction with the user with code like this:private async Task MessageReceivedAsync(IDialogContext context, IAwaitable<IMessageActivity> result) { PromptDialog.Choice( context, this.AfterChoiceSelected, new\[\] { ShowMyTeamsOption, CreateTeamOption }, "Hi I'm Theodor your Team automation but. How can I help?", "I am sorry but I didn't understand that. I need you to select one of the options below", attempts: 2); }This will create our two different buttons and will guide the user in the directions we want them to choose from. After we have all the data from the user, in our case just an alias and a display name, we can now call the Azure Function handling our request.``` Dictionary<string, string> dictionary = new Dictionary<string, string> { { “displayName”, this.displayname }, { “alias”, this.alias } };

Speaking at Collab365 Conference: Raspberry Pi, Azure & Office365

Fri, 20 Oct 2017 13:38:03 +0000

Have you heard about the virtual Collab365 Global Conference 2017 that’s streaming online November 1st – 2nd? Join me and 120 other speakers from around the world who will be bringing you the very latest content around SharePoint, Office 365, Flow, PowerApps, Azure, OneDrive for Business and of course the increasingly popular Microsoft Teams. The event is produced by the Collab365 Community and is entirely free to attend. Places are limited to 5000 so be quick and register now. During the conference I’d love you to watch my session which is called : ‘Raspberry Pi, Azure & Office365 - How MoCaDeSyMo informs us about lunch’ The session covers the story of MoCaDeSyMo, a Raspberry Pi taking pictures and uploading them to Azure. With the power of the Azure Custom Visions Service, based on a trained image recognition model we are able to tell if the local food truck is already here or not. MoCaDeSyMo than uses Microsoft Teams to inform the different subscribers. For further reference, all the information is also assessable through a SharePoint Online site. If you join me, you will learn:

Raspberry Pi + Azure Functions + Cognitive Services + Microsoft Teams -> How a Pi informs us about lunch

Mon, 26 Jun 2017 21:08:34 +0000

Introducing MoCaDeSyMo

Like every other office, ours also has different opportunities for lunch. You can take your break and visit the cafeteria, you can order food from some local restaurants or you can just step outside and buy something from the food truck. I’m a regular food truck buyer, but this option has some downsides. My office is on the other side from the front door, so I don’t see the guy arriving. This means I end up “pinging” colleagues with a direct line of sight regularly in Skype for Business in order to get some information about when I should go for lunch. Right now there are only a few people in our team with a window facing the front door so they end up getting pinged a lot. To end their misery we came up with the idea of some type of push notification. The first idea was about writing a mobile app that the driver of the food truck installs on his phone. This app should recognise our GPS coordinates and push a simple message to Azure upon arrival at our front door. This could have worked but had a dependency on the driver’s phone and his free will to join the idea. After some further thinking, we came up with the idea of using a Raspberry Pi that takes pictures of the scene and uploads them and uses cognitive services to check whether the food truck is present or not. So without further ado, here is MoCaDeSyMo: MoCaDeSyMo(mobile carbohydrate delivery system monitor) is the red fellow on the left of the picture. Insides his red body is a Raspberry Pi 3 and the default Pi camera module. On weekdays between 10:45 and 11:30 MoCaDeSyMo takes a picture every minute and tries to detect the food truck of a local snack bar called “Imbiss Rainer”. His main target is to recognise something like this: To make sure everyone on the team knows about the arrival of the truck MoCaDeSyMo uses Azure Cognitive Services to predict the presence of the food truck and if the probability is above a given threshold a Microsoft Teams connector gets triggered and we get notified with a message like this: This post discusses how we created the system based on this high-level architecture:

Wie man einen Q&A Bot für Microsoft Teams erstellt ohne eine Zeile zu coden

Mon, 13 Mar 2017 21:53:15 +0000

Vor knapp einer Woche, am SharePoint Saturday in München, hielt Wictor Wilén eine spannende Session zum Tema Microsoft Teams. Er zeigte dabei die Möglichkeiten auf mit denen man Microsoft Teams erweitern und um Tabs und Connectoren ergänzen kann. Zusätzlich nutzte Wictor auch die Gelegenheit seinen neuen Yeoman Generator für Tabs in Teams zu veröffentlichen (http://www.wictorwilen.se/yo_teams-tab).

Inspiriert von diesem spannenden Vortrag war es an der Zeit selbst zu versuchen in wie weit man das dort gezeigte auch bereits im eigenen Tenant verwenden kann. Ziel war die Anlage eines QnA Bots ohne dafür eine Zeile Code zu benötigen und genau das wollen wir uns nun genauer ansehen.

SharePoint Wiki Bot - How to automate the Azure QnAMaker bot knowledgebase

Sun, 12 Mar 2017 19:04:46 +0000

<TL;DR>

This post shows a proof of concept implementation of an Azure QnAMaker bot that gets its knowledgebase updated by changes in a SharePoint wiki. The main idea is, that you use a wiki as some sort of knowledge management system and that changes to the content automatically get pushed to the bot’s knowledge base. By adding a SharePoint Webhook to the wiki pages library you can trigger an Azure web job that manages the bot’s knowledge base. By using the Azure QnAMaker API the web job publishes changes to the knowledge base so that a connected Azure bot can use it as it’s backend data. This way you can add a heading and some text to the SharePoint wiki page and about two minutes later you can ask the bot in Skype or Microsoft Teams about it. Kudos to: Wictor Wilén, Bert Jansen and Paul Schaeflein.

How to create your own Q&A bot for Microsoft Teams without code

Sun, 05 Mar 2017 18:46:37 +0000

Introduction

Yesterday at SharePoint Saturday Munich I had the pleasure to attend Wictor Wilén’s session on extending Microsoft Teams. In this session Wictor showed some great demos and even announced the availability of a Yeoman generator for creating new Tabs in Teams (http://www.wictorwilen.se/yo_teams-tab).

Wictor’s talk inspired me to give the Microsoft Bot Framework a try and so I started creating a bot in my Azure tenant by the guidance of this blog post:Create BOT from FAQ page with no coding — QnA Maker as new member of Cognitive Service.