Pål Bendiksen
Cloud Solution Architect and technical mentor with focus on Azure and Enterprise Integration
Oslo, Norway
Actions
PB has been in the industry for more than 20 years. His work experience includes companies like Universal Music, Deloitte, DNV and 10 years at Microsoft. Since 2010 he has worked on cloud solutions, focusing on hybrid cloud and integration. Working at Microsoft, Azure naturally became the cloud of preference, but he has great passion for technologies that enables systems to cooperate across different cloud providers. His primary focus is his delivery to client and mentoring colleagues.
Area of Expertise
Topics
Beyond the Toy Phase: Harnessing the Power of AI for Business Impact
Artificial Intelligence (AI) has become an integral part of our daily lives, from personalized recommendations to voice assistants. But how can businesses truly harness the power of AI, specifically large language models (LLMs) like OpenAI's ChatGPT, Bing Chat, and Google Bard, to generate tangible return on investment (ROI)? In this engaging talk, we will delve into the practical applications of LLMs in the business world, focusing on driving value and managing associated challenges.
LLMs, comprised of enormous datasets, hold incredible potential, but only a small portion of their knowledge may be applicable to your specific business needs. How can we extract valuable insights from these vast resources? We will discuss strategies for employing LLMs effectively in enterprise contexts, maximizing their relevance, and deriving meaningful ROI.
Our session will feature a live demo showcasing an AI-driven integration project, in which we implemented an LLM-based Teams Chat Bot to streamline information sharing across organizational departments. Witness how effortless access to company-wide data can lead to increased efficiency and improved decision-making, benefiting various stakeholders.
We will also address potential pitfalls and risks associated with LLM utilization, underscoring the importance of striking a balance between leveraging AI for business growth and mitigating exposure to unforeseen challenges.
Join us as we explore the untapped potential of LLMs, sharing practical insights and examples to help businesses unlock the true power of AI and revolutionize their operations. Don't miss this opportunity to gain competitive advantage in the ever-evolving world of technology.
Securing your Azure integration platform
Most systems that have been around for a while has at least some form of integration with other services, software or systems around it. If your IT platform is a bit mature, you might also have an integration platform - a central hub that handles all your integrations and data flow.
But how about security? Your data is flowing around between all these systems and services, but who and how is it ensured that your precious data doesn't end up in someone that's not authorized to access it? Do you ensure that all systems only see what hey need, and how do you handle this?
Are you using passwords or API keys to secure your endpoint? And how does your systems exchange data between them on the "back side"? It may be hard to see where your back-end systems are more accessible than what you actually wish for.
The services in Azure is easy to set up. API Management, set up using some Logic Apps or Functions to do the data load and processing. But hos secure is it? Can you access your Functions omitting the security you built in to the API management?
The security isn't necessarily built in "out of the box", and you really need to consider this - and the earlier in the process the better. You can end up with your back-end systems being wide open to the world, and not even see the unauthorized access in your logs, because you are logging on the wrong level (the level where security actually is handled).
In this session we'll take a deeper dive into how to handle security and network isolation in an integration platform deployed in Microsoft Azure. The principals will absolutely be transferrable to other cloud platforms as well.
We'll discuss the implications of complexity and pricing, what considerations and trade-offs you might have to look into. All taken into consideration your project's budget and data sensitivity.
We'll look in to three main models, ranging from full network isolation to no network isolation. The lesser network isolation, the more work and planning has to be put down into how to secure your platform and your data.
We will guide you through some of the mistakes, so you don't have to make (and correct) them in your own projects, some of the most important pitfalls and some of the biggest cost drivers.
Azure Logic Apps in the Enterprise
Logic Apps is a great tool for POC's and one-off data moves, right?
Yes, it is!
But Microsoft did not put all that effort into a tool that is not intended for production scenarios.
Logic Apps is an Enterprise Application integration tool that can be used to create on-demand, event based and scheduled integrations. You can even create APIs to expose data from your data sources in a matter of minutes.
But how do you ensure that ensure that you meet all the enterprise requirements like:
• Performance
• Security
• Monitoring
• Operations
• Cost
• Support for Legacy systems and protocols
Be aware that this does not come as a part of the package by default. It is way easier to miss these requirements than in other services that require more developer experience. If you are not careful, you will end up with an Enterprise solution with POC-like performance, security and with no way to monitor attacks, application errors, usage etc.
We will show you how you can make all the mistakes in a matter of minutes :)
We will of course show you how you can avoid some of the mistakes and create solutions that are Enterprise ready using this fantastic service.
Some of the things we will go through:
• Several ways of avoiding loops to increase performance and reduce cost
• Why data operations is you best friend
• Handling XML, ways of converting to json, and why some of them just doesn't work
• Transformation with and without Liquid templates
• Differences between Logic Apps Consumption tier and Logic Apps (Standard)
• Intro to isolating Logic Apps
• Using managed identity to increase security
• Monitoring and operating multiple Logic Apps
• Connecting to legacy systems
• Decoupling cloud systems to simplify operations
• Creating an API with asynchronous response in less than 2 minutes
We will also touch upon services like Api Management, Data Factory and Service Bus
Securing your mission critical cloud workloads
Most companies have now started using the cloud for running some workloads in their organizations.
The first workloads to come to the cloud is usually non-critical, non-sensitive workloads like campaign websites and other public websites. The impact in case of a succesful attach is very low, and the security is often up to the developers.
Now more and more business critical highly sensitive workload are moving into cloud.
So how do we handle this, when the workloads that used to be hosted on premises under the safe care of the IT staff that are experts on network security and logging all suspicious behavior?
This will be a session about how we do this in Azure for some of the most critical customers in Europe. But many of the concepts will apply top other clouds too.
Come by for a session about Landing zones, network isolation, deployment patterns and security, Key vaults, authentication and much more related to architecting and developing secure cloud applications.
API World 2023 Sessionize Event
CodeCampSDQ 2023 Sessionize Event
Caribbean Developers Conference 2022 Sessionize Event
Internal microsoft events
Presenting session at several internal events (TechReady) at Microsoft hosting 5000-6000 participants
Climate Hackathon 2022
Winner of Climate hackathon for Building Energy Level Predictor. https://www.youtube.com/watch?v=Hln9121Uqwc
Sangam 21
Remote Session: Common Multicloud Scenarios and Best Practices with OCI
https://www.aioug.org/sangam21-sessions?session_id=267
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top