Jason Monsorno
Director of Engineering
Jacksonville, Florida, United States
Actions
Microsoft MVP with over 15 years of experience developing and refining AI systems across model engineering, agent architectures, and large‑scale platform design. Recognized for turning complex technical concepts into clear, actionable designs that enable developers to build robust, reliable solutions.
Links
Area of Expertise
Topics
Turning LlamaSharp into a Private, Offline‑Capable AI Agent
Explores the example evolution of LlamaSharp from a simple inference library into a fully local AI agent. The session details the architectural components that enable this shift, including prompt‑routing mechanisms, contextual memory models, tool‑integration workflows, and controlled system‑interaction layers. It focuses on the engineering patterns and design considerations required to support autonomous behavior while maintaining strict on‑device execution and privacy.
Device Code Phishing: The Entra ID Attack Vector Hiding in Plain Sight
The device code authentication flow was designed for input-constrained devices like smart TVs and kiosks — but attackers have quietly turned it into one of the most effective ways to steal valid Microsoft 365 tokens without ever touching a password. Unlike credential phishing, device code attacks bypass MFA entirely, leave minimal log noise, and work against any tenant running default Entra ID settings.
In this session, we'll break down exactly how device code phishing works: from the initial lure email to the moment an attacker walks away with a persistent, legitimate access token scoped to your tenant. You'll see a live walkthrough of the attack chain, understand why conditional access policies alone won't save you, and learn what the attack looks like in Entra ID sign-in logs — because knowing what to hunt for is half the battle.
We'll close with a practical hardening checklist covering authentication flow restrictions, Conditional Access controls, continuous access evaluation, and how Microsoft's new token protection features fit into your defense strategy. Whether you're an admin, a security architect, or a defender building detections, you'll leave with specific, actionable controls you can implement the same week.
Optimizing AI Usage
Most people interact with AI the same way every time — typing similar prompts, running through the same steps, and burning through tokens on tasks that could be automated or built once and reused. This session reframes how you think about AI: not as a chat partner you re-engage from scratch each day, but as a platform for building durable tools, automations, and programs that do the work for you on an ongoing basis. We'll explore how to identify repetitive AI workflows and convert them into scheduled tasks, persistent artifacts, or lightweight programs that run without manual prompting. You'll learn when it makes sense to have AI write a script or build a tool versus just answering a question — and how that shift can dramatically multiply your output without multiplying your effort. Walk away with a smarter approach to AI investment: one that compounds over time instead of resetting every conversation.
Streamlining MCP Tool Delivery Through dnx
Explore how dnx streamlines the development and deployment of MCP tools by simplifying packaging, distribution, and rapid local iteration. The session focuses on the core mechanics with real world examples to provide consistent tooling behavior and smoother integration within MCP‑based agent ecosystems.
Managing Remote Workers Effectively
As remote and hybrid work becomes the norm, managers face a new set of challenges — from maintaining team cohesion across time zones to keeping individuals engaged without face-to-face interaction. This session explores practical strategies for setting clear expectations, building trust, and creating accountability structures that work outside a traditional office environment. We'll cover communication best practices, how to run productive virtual check-ins, and how to spot early signs of burnout or disengagement before they become bigger problems.
Not Just Another API: Understanding MCP and Why It Changes Everything
APIs have been the backbone of software integration for decades — but AI systems have different needs. The Model Context Protocol (MCP) is an open standard that lets AI agents connect to tools, data sources, and services in a structured, consistent way. It looks familiar on the surface, but treating MCP like a conventional API leads developers down the wrong path fast. This session covers what MCP is, how it differs from REST and function-based APIs in terms of state, discovery, and intent, and why those differences become critical the moment an LLM is making decisions autonomously. We'll also touch on getting started — connecting to an existing MCP server and walking through real-world examples that show what's possible once everything is wired together. Attendees will leave with a clear mental model of where MCP fits, when to reach for it over a traditional API, and a concrete starting point for their first integration.
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