73 зустріч Microsoft .NET User Group Odesa

Важлива новина! В п'ятницю 3 листопада в 18:30 в тімсі відбудеться 73 зустріч нашої групи! Буде супер доповідь та, як завжди спілкування після неї!

Чекаємо на вас! Детальний опис доповіді нижче.

Посилання на ІВЕНТ.

Example of how to bring GenAI RAG architecture to production.

We are going to dive a bit into a world of Gen AI, get familiar with main concepts and in detail will look at RAG architecture as one of popular approaches in the field. I'm going to show you an example of a production ready web API application based on python library FastAPI that implements RAG architecture as well all software engineering best practices. We will currently touch hyped frameworks and technologies such as Llama-Index, Langchain and GPT, Microsoft Semantic Kernel. Im happy to answer all questions.

Alexander Demchuk, Solution Architect AWS/ Gen AI at Provectus

72 зустріч Microsoft .NET User Group Odesa

Важлива новина! В середу 27 вересня в 18:30 в тімсі відбудеться 72 зустріч нашої групи! Буде супер доповідь та, як завжди спілкування після неї!

Чекаємо на вас! Детальний опис доповіді нижче.

Посилання на ІВЕНТ.

Building an architecture for a modern GenAI Application by Anton Vidishchev

GenAI apps are hyping, but in many cases it is not clear how to combine different best practices to build a proper architecture.

We will look into the architecture of a hackathon-winning project, Quizzilicious.com, and discuss how to put together various must-have practices, such as metaprompting, retrieval-augmented generation, responsible AI, and some others. This includes looking into the infrastructure and project code!

Anton works as a lead architect in Capgemini Engineering. He has over 20 years of experience in software development, mostly specializing in Microsoft technologies and Azure. Lately he is one of the drivers of applying GenAI in Enterprise projects. He is also a passionate conference speaker and one of the organizers of Odesa .NET Community.

Global Azure Odesa(Ukraine) 2023

Група повертається онлайн після року війни. Зустрічаємось в суботу 13 травня в 19:00 в тімсі щоб послухати дві шикарні доповіді від топових MVP з України та Франції. А також поспілкуватися після доповідей, дізнатися як справи та побачитись хоч віртуально.

Чекаємо на вас! Детальний опис доповідей нижче.

Посилання на ІВЕНТ

1) Secure your app in under 60 minutes. By Anton Boyko

The war started unexpectedly, and many web apps were unprepared for that. While soldiers were fighting in the physical world, the hackers were fighting in the cyber world.

This session is based on my experience during the first days of the war. A non-commercial organization reached out to me asking to protect their web app, their users, and their data from cyber invaders.

So, what can you do in 60 minutes working on the mobile internet from the basement while missiles fly over your head?

Anton Boyko - 20 years in IT industry. Worked as Developer, DevOps and Architect. Participated in global worldwide projects with millions of users load. Currently occupies the role of the Chief Infrastructure Officer at Perfection.DEV. Has his own IT consultancy practice, helps organizations of different sizes to be more efficient by investing in modernization. Microsoft Azure MVP since 2014. Microsoft Regional Director since 2020.

2) Deceptive Opinion Spam Detection with Azure Machine Learning. By Alibek Jakupov.

Deceptive Opinion Spam commonly takes the form of fake reviews (negative or positive) posted by a malicious web user to hurt or inflate a company’s image. As these reviews have been deliberately written to deceive the reader, human reviewers are faring little better than a chance in detecting these deceptive statements. Thus, there is a dire need to address this issue as extracting text patterns from the fraudulent texts with meaningful substructures still remains a challenge. In our presentation, to obtain a deeper understanding of how lies are expressed in texts, we we constructed several models with Azure ML to learn the patterns that constitute a fake review, and then explore the outputs of this model to identify those patterns. As the linguistic cues of the lies are still unknown, a key advantage of out approach is that we make the representations more interpretable and provide additional opportunities to reveal the patterns and structures within the systems of documents.

Alibek Jakupov - Lead software engineer at Expertime France, with 6+ years of experience as Data Scientist and ML engineer, Microsoft AI Most Valuable professional since 2019, author of several scientific publications.