AI in production. What enterprises have actually built, how it is working, what it costs, and what breaks in the real world.
MCCD26 is presented by
THE FORMAT
A closed-door conversation among peers
MCC Day is a curated, invitation-only event. No recordings. No product pitches. Under 100 executives and technical decision-makers from financial services, telco, banking, and enterprise, together for a full day of talks, confidential roundtables, and an evening on the rooftop terrace.
Curated
Every seat considered. Fewer than 100 attendees. The room is a feature of the event.
Confidential
No video, no recording. What’s said in the room stays there. That’s how you get the real numbers and the real trade-offs.
Peer-led
Talks by the people who built and ran the systems. Not the people who sell them.
THE FORMAT
Produced in partnership with:
The content
AI in production. What is next?
The theme is AI in production: what it costs, what it means to secure it, and what changes about software development and operations.
session 1A
Weaponized AI
The new attack surface of the agentic era | Prof. Stefano Zanero, Politecnico di Milano
AI in production no longer just talks: it books, integrates, pays. Every new capability is a new attack surface, and most share one root flaw: models treat content as instructions. A map of the agentic threat landscape, from prompt injection through MCP tool poisoning to attacks on agent payments.
Deterministic guardrails for non-deterministic intelligence | Andrea Cavezzale, Moviri & Rachid Es-Skidri, Moviri
As AI agents evolve into autonomous execution engines, their speed creates serious attack vectors like prompt injections and memory poisoning. Harnessing their efficiency safely requires deterministic software guardrails around non-deterministic AI. Combining strict ingestion hygiene, execution middleware, and code-enforced human governance allows organizations to scale high-velocity automation while insulating core assets and maintaining full operational control.
Red-teaming your enterprise AI stack | Piero Tallarico, Moviri
Jailbreaking the model is the easy part. Real exposure lives in the full stack: RAG pipelines, vector databases, orchestration, tool execution. A practical blueprint for offensive testing of AI systems, including automated red-teaming with adversary models, open-source tooling, and the hardening measures that actually work.
AI systems for autonomous fraud operations | Nicolò Pastore, Cleafy
Attacks now assemble themselves: a model writes the code, a command-and-control server reads victims’ messages and picks the targets, a call module runs the conversation with nobody on the line. This session sets out the operating architecture that replaces it: continuous observation across both perimeters, external and internal. An AI-native operating system for the fraud unit, specialised agents carrying fraud knowledge, with a human on the loop.
session 1B
Sovereign AI
Sovereign AI: Is building an EU alternative possible? | Charlie McGhee, Scaleway & Davide Brunelli, Moviri
Sovereign AI has moved from political slogan to business urgency: European enterprises know it, but most still don’t have a plan. Relying on non-EU technologies is no longer just an ideological question — it’s a concrete risk to operational continuity, cost control, and strategic autonomy. The window to act is closing fast. What’s needed now is a practical path to building technological independence across the entire digital stack — before Europe decides it for you.
Beyond open models: When AI sovereignty is worth the investment | Federico Sallemi, ContentWise
When does training your own model make sense? Sharing a customer’s experience of building a foundation small language model from the ground up: the sovereignty and domain drivers behind the decision, what it took in data, compute, and people, and what they’d do differently.
Building your own foundation model in-house | Enrico Mannarino, Moviri
When a company decides to build its own model, navigating the “open” world is complex, as it offers resources but no complete recipe. This talk explores a concrete case: training an 8-billion-parameter model from scratch for internal use. It covers the essential steps – data preparation, training stages, cost and schedule drivers, post-training, and evaluation – providing practical advice before you start.
Custom vision models: Building vision AI for the real world | Paolo Cremonesi, Moviri & Politecnico di Milano
Language models get the headlines, while vision models power real-world applications such as inspection, safety, document processing, and monitoring. Custom and foundation vision models enable a wide range of visual tasks through adaptation and fine-tuning, while posing challenges in training, deployment, and lifecycle management. Within a sovereign AI architecture, they provide greater control over data, infrastructure, and production workloads.
session 2a
Agents at Work
The agentic IT Ops stack | Anna De Angeli, Moviri
Ops teams spend most of their resolution time on investigation, not fixing. Anomaly detection isn’t the bottleneck anymore; reasoning is. How multi-agent systems take on the full incident lifecycle, from context-enriched triage and root cause analysis to remediation, and what infrastructure autonomous workloads require.
When agents run the SOC: Rethinking security operations | Samuel Venturini, Moviri
Adding AI to a broken SOC process just produces faster noise. This talk treats security operations as a process design problem: which workflows should be rebuilt around autonomous triage and investigation, where humans stay in the loop, and what changes for SIEM when agents are the primary log consumers.
Accelerating scientific research with AI agents | Filippo Lunghini, Dompé & Michele Scapinello, Moviri
Building reliable AI agents for scientific discovery requires more than a powerful model. A biopharma R&D case: exploring how trusted knowledge, intelligent orchestration of research capabilities, evidence-grounded explainability, and systematic hypothesis validation come together to enable AI-assisted scientific discovery.
Agentic coding: Real-world lessons from replatforming projects | Mattia Zeni, Databricks & Manuel Salamino, Moviri
Agentic coding tools promise order-of-magnitude productivity. Replatforming projects are where the promise meets legacy reality. What worked and what didn’t when agents took on real migration work: where they excel, where they fail silently, and how team structure and review practices had to change.
session 2b
Token Economics
Exponential AI adoption: What will be scarce? | Giovanni Bonfanti, EY & Paolo Bozzola, Moviri
A framework for the numbers that matter: cost per task, inference routing, fine-tuning vs. prompting breakeven, and the build/buy/bypass decision. How to treat token budgets as an architectural constraint rather than a billing surprise.
Cracking the AI inference stack: Autonomous tuning for maximum AI efficiency | Stefano Doni, Akamas
We are entering an era of global compute scarcity where AI efficiency is the difference between million-dollar savings and lack of capacity for business growth. Join us to demystify the modern AI inference stack performance. We’ll share our lesson learned using AI autonomous tuning to optimize AI full-stack configurations, delivering 50%+ tokens throughput and 2X faster interactivity, to squeeze maximum ROI from your AI infrastructure.
Beyond LLMs: Building a global AI grid at the edge | Federico Navarra, Akamai
Centralized inference has a physics problem: latency, egress costs, and data gravity. Akamai shares how a globally distributed edge compute grid changes the economics of AI serving, which workloads belong at the edge, and what an inference architecture looks like at planetary scale.
The convergence of finops and tokenomics | Roberto Martini, Moviri
Unlike traditional cloud costs based on fixed provisioned capacity, AI expenses fluctuate dynamically based on system execution and token consumption. Managing this shift requires adapting FinOps practices to variable token-based pricing.
All MCC Day sessions are in Italian. MCC Day will be followed by the annual Moviri Terrace Party.
THE FORMAT
Join a group of industry leaders
A small, curated group of senior technology and product leaders from European financial services, telco, banking, and enterprise organizations. Decision-makers who are shipping AI systems, or accountable for the ones their organizations are about to.
3rd
20+
<100
1
PAST ATTENDEES
They joined us
THE location
September 17, Milan.
Moviri HQ, Via Schiaffino 11
schedule
Core agenda 14:00 to 18:00 · Optional lunch and aperitivo
format
Talks · Roundtables · VIP Happy Hour · Evening Party
access
Invitation only, subject to host approval
Seats are limited and reserved for a select group of peers. Attendance is subject to approval.
