Edoardo Dusi
Senior Developer Advocate
Senior Developer Advocate
Bologna, Italy
Actions
Edoardo Dusi is a Senior Developer Advocate at AWS, based in Italy. He started as a software developer, moved into leading engineering teams, and then into developer relations, where he ran the DevRel & Community team at Storyblok before joining AWS. He shares what he learns through talks, his blog and Buongiorno da Edo, a tech and AI podcast on YouTube that he writes and hosts himself.
Edoardo Dusi è Senior Developer Advocate in AWS e vive in Italia. Ha iniziato come sviluppatore, è passato a guidare team di ingegneria e poi alla developer relations, dove ha guidato il team DevRel & Community di Storyblok prima di entrare in AWS. Racconta quello che impara attraverso talk, blog e Buongiorno da Edo, un podcast su tech e AI su YouTube che scrive e conduce in prima persona.
Badges
Area of Expertise
Topics
If MCP is stateless, is it just REST?
The new MCP specs from July 2026 removed sessions, the handshake and Mcp-Session-Id are gone. The question you may be asking after looking at that is: is it just REST...with extra steps?
In this talk I will try to answer with data, exposing the same backend in a Kubernetes cluster as a stateless MCP server and as a REST API backend, giving both to the same agent with the same tasks. We can then measure what changes for the agent, like the success rate and the tokens spent, and how each asks the user for missing input or instructions. Then we can see what changes for the humans, the platform team, and where REST still wins (spoiler: HTTP caching).
You will leave with a better understanding of when to use stateless MCP when you need your agent to discover (and self-correct against) tools, and when to use REST when your caller is code and you want caching and a clear, cache-friendly contract.
Honey, We Predicted the Bits
80s sci-fi movies gave us rays that miniaturized humans and things without losing a single atom. That's what we call "lossless" compression in computing, where we want to shrink web assets, code, and payloads where 100% bit fidelity is non-negotiable.
As an article that circulated a lot this summer states, compression is prediction. The better a system can predict what comes next, the fewer bits we need to transmit. But we have real engineering constraints!
In this talk, we will explore:
- The classical and modern web stack, how we evolved from basic redundancy to today's landscape
- Why the model must be part of the message
- The format-aware frontier, so how OpenZL uses DAGs, structural knowledge and universal decompressors
- The connection with LLMs, where the prediction of the next token and the cross-entropy loss are mathematically identical to optimal entropy coding... and why language models are great predictors, but impractical compressors
Design your PHP app so AI agents can drive it safely
AI agents are now operating on real web applications, not just helping you build them. So for a team working on a new or existing PHP app the question is no longer "which assistant should we adopt" but "is our app ready to be driven by an agent"?
Agent-readiness is a design problem, and PHP devs already have the skills to solve it. In this session w see how to expose a Laravel app via an MCP server, turning operations into a small and constrained set of enumerated tools that an agent can call safely, with structured output and scoped permissions.
A working agent built on the open source Strands SDK and Amazon Bedrock, demonstrates the design end to end: it discovers the available tools, reasons about a request, and acts on the app.
Total Recall. Get ready for a surprise.
You followed the tutorials. You embedded your docs, wired up a vector database, and shipped a shiny RAG chatbot...only to watch it confidently invent facts. The standard playbook is letting you down.
The twist? The wins in production are often simple and counter-intuitive. The secret isn't a bigger model, it's smarter retrieval.
This talk reveals the overlooked patterns that turn a POC into a production-ready system:
Surprise #1: Your chunks are wrong.
Stop naïve text splitting. Use your content's native structure (headers, JSON) to keep context intact in your embeddings.
Surprise #2: Keywords still matter.
Hybrid dense+sparse (BM25) raises coverage and precision, especially for entities, IDs, and exact phrases.
Surprise #3: Retrieval is two-stage.
Add reranking with a stronger cross-encoder to sift the true signal from top-k noise.
Surprise #4: You can’t steer blind.
Run evals (e.g. Recall@k) so iterations move the numbers, not just the "vibes."
No RAG 101. We'll show before/after metrics, failure modes, and the trade-offs that separate prototypes from production systems. You'll leave with a pragmatic playbook for retrieval that actually remembers.
Attendees leave with a concrete playbook for high‑fidelity RAG retrieval.
I'll (try to) teach developers how to design smarter retrieval pipelines and prove improvements with metrics. The focus is on structure‑aware chunking, hybrid dense+sparse search, two‑stage retrieval with reranking, and evals that guide iteration beyond vibes.
Honey, I Shrunk the Bits. Or how modern HTTP compression works.
In today's world of complex web applications and micro-frontends, understanding HTTP compression is about more than just sending smaller files, it's crucial for creating better user experiences. This technical overview looks at how modern compression algorithms are influencing the way we build and deploy web applications.
We will start from the basic understanding of compression algorithms, lossy and lossless, and their foundational strategies, then have a look at the modern compression landscape with Gzip, Brotli and Zstd, to see how they work internally, their similarities and differences, and how we should leverage them for our web applications.
Technical Requirements:
Understanding of HTTP and asset delivery
Experience with modern JavaScript frameworks
Basic knowledge of web performance metrics
Target Audience:
JavaScript developers
Performance engineers
DevOps engineers
Framework architects
Session Duration: 35 minutes + Q&A
Format:
Technical deep dive with visual explanations
Real-world performance measurements
Framework-specific examples
Interactive compression analysis demonstrations
Recommended Conferences:
JSConf
Performance.NOW()
Next.js Conf
React Summit
ViteConf
Angular Nation
Key Takeaways:
Attendees will understand:
How different compression algorithms affect JavaScript delivery
Best practices for compression in modern web applications
Framework-specific compression strategies
Performance monitoring and optimization techniques
Supporting Materials:
Compression analysis toolkit
Configuration templates for various hosting platforms
Benchmark suite for compression testing
Devfest Pescara 2026 Sessionize Event Upcoming
Codemotion Milan 2026 Sessionize Event Upcoming
Øredev 2025 Sessionize Event
reactjsday 2025 Sessionize Event
DevFest Modena 2025 Sessionize Event
Come To Code 2025 Sessionize Event
WeAreDevelopers World Congress 2025 Sessionize Event
phpday 2025 Sessionize Event
Codemotion Rome 2025 Sessionize Event
DEVWorld 2025
Deep dive into JavaScript Bundlers
Come To Code 2024 Sessionize Event
WeAreDevelopers World Congress 2024 Sessionize Event
DeveloperWeek Global 2024 Sessionize Event
Open Source Day 2024 Sessionize Event
WeAreDevelopers Live 2024 (Season 7) Sessionize Event
dxday preview @127.0.0.1
My session: Backstage: Platform Engineering Revolution powered by Open Source
Devfest Alps 2023 Sessionize Event
Devfest Pescara 2023 Sessionize Event
reactjsday 2023
React vs Rest of the World
A technical comparative between react approaches and patterns and other frontend frameworks
containerday 2023
My session: WebAssembly: The Next Frontier of Cloud Computing
containerday is the leading Italian conference dedicated exclusively to containerization and virtualization technologies and associated best practices.
Come to Code
Clash of JS: the same app built with different JavaScript frameworks
WeAreDevelopers Live 2023 (Season 6) Sessionize Event
React JS Milano Meetup
Testing web apps with Cypress
WMF - We Make Future 2023
DevRel program, strategy and KPIs in a service company
DeveloperWeek Europe 2023 Sessionize Event
WeAreDevelopers Live 2023 (Season 5) Sessionize Event
Web Day 2023 Sessionize Event
angularday
angularday 2022 is the 6th edition of the Italian angular conference, organized by GrUSP, organizers of events like phpday and jsday.
My session: https://2022.angularday.it/talks_speakers/#testing_cypress
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