Why not start a new year by learning something new? I’m visiting Microsoft Ignite 2020 — let’s see what’s there to discover this time. For readers that have not been to this event yet, it’s an annual conference happening in different cities all over the world led by Microsoft experts and the community. Usually you choose a learning path, depending on what you’re interested in, build your schedule and go learn something new. There are presentations for people with all different backgrounds, from basic introduction to cloud and discovering tools and features available, all the way up to expert talks about dev-ops, saving costs for cloud services and building your next cloud business solution. Make sure to pre-register as early as possible — the event is free, but there’s a limited amount of spaces available.
Since there weren’t too many technical talks this year, I chose the “Developers guide to AI” path. Microsoft is trying to push Artificial Intelligence and Machine Learning technologies to the limit and make user experience as smooth as possible. From building machine learning models without code to deploying pre-built AI to live servers, chat bots, speech recognition, business intelligence, future predictions, computer vision, you name it, it’s all ready and available on cloud.
- All possible paths can be found on the Microsoft Ignite the Tour website.
Some of the paths contain all of the material presented, but not all of the presentations are uploaded to the site. Usually if you are at the presentation the speakers present an extra slide with a link to all the material at the beginning and end of the talk. From my experience it’s a way easier and more reliable to find the relevant materials that you’re interested in from these.
Developers guide to AI
I’ll go into a bit more details and a short summary for what was presented on each of the AI topics but make sure to check out the path site for more details and in-depth samples or presentation slides. For most of the presentations Tailwind Traders business samples were used. It’s a fictitious retail company showcasing the future of intelligent application experiences. The project is open source and you can use it for building your next business solution or running some tests.
- You can find all the reference samples on the Microsoft TailwindTraders github.
- All the Developers guide path presentations are available on the Microsoft Ignite the Tour website.
Making sense of your unstructured data with AI
Presentation on how you can use Cognitive Search to make sense of both structured and unstructured data. Discussion about AI concepts, like the ingest-enrich-explore pattern, skillsets, cognitive skills, natural language processing, computer vision, and others.
Using pre-built AI to solve business challenges
Really interesting and practical presentation with some demos on how to use AI to improve product search on the Tailwind Traders site. If you’re missing a piece from under your sink and don’t even know what it’s called, why not take a picture and use AI to search for it. Check out this talk if you want to know how text to speech, photo search, user behaviour analysis or other pre-built AI models can help out your business.
How to build machine learning models with no code
Building custom Machine Learning Models is not an easy task and requires a lot of effort, or does it? In this presentation you can learn how to build ML model right on Azure portal without even having to do any coding. Choose data sources, train and test the model and eventually deploy everything to an API.
Taking models to the next level with Azure Machine Learning best practices
AI and ML can greatly increase business processes productivity, but sometimes existing models are just not enough. A session on how to train your own model for more specific tasks or how to increase its accuracy. In some cases it’s even hard to identify what sort of model can fit your business needs, that’s where AutoML can help out creating a model from scratch.
Machine learning operations: Applying DevOps to data science
In some cases, collaboration between developers and data scientists can be very limited and deploying ML models to production in a consistent way is often a pipe dream. An in-depth session on how to apply DevOps practices to ML projects using Azure DevOps and Azure Machine Learning Service.
The conference overall was a really great experience despite the huge amount of people that came over to visit this year it was really well organised. Glad to see Microsoft taking community reviews into account and making adjustments on how the talks are spaced out to have at least a short break between them so you don’t have to run from one end of the venue to the other just to make it on time to see another session. Make sure to check out more details about any sessions you’re interested in and see you in the next Microsoft Ignite conference!
