About Us
We Are Prévenance
Built Differently. Deliberately.
Remote-First
You are not paying for our postcode or our city-centre offices. When a large firm flies a team of analysts to your company location for a two-week on-site, you are funding that infrastructure whether you see it on the invoice or not. We start the moment you grant us secure access to your systems.
AI-Native
We operate at a speed human teams cannot match. Every output Prévenance produces passes through a Human-in-the-Lead (HITL) verification layer: senior experts who apply contextual judgment, remove false positives, and ensure that what the AI surfaces is tactically sound before it reaches you.
Discretion as a Structural Feature
Prévenance can operate within your existing infrastructure, using the same tools your teams already use. To our clients, we present as a well-organised internal function not an external team parachuted in to assess them.
Engage Prévenance
Meet our founder
Shannon Sedgwick
Shannon is a 15-year technology operator. He founded GM Risk Group in 2012, scaled it to eight-figure ARR across offices in San Diego, Sydney, and Singapore, and exited at 1.5× AR in 2018. He subsequently led P&L-owning roles at MinterEllison (Partner, national technology practice), Ankura (SMD, APAC — US PE-backed), Verizon, and Deloitte. Alongside his operating roles, he has chaired AI SaaS company Castlepoint Systems through capital raising, directed M&A due diligence on $100M+ transactions, and served on multiple boards. He holds an MBA, a Master of Cyber Security, and is a Graduate of the Australian Institute of Company Directors (GAICD).
FAQS
Frequently Asked Questions
Data Leakage
Public AI tools are "open loops". Any data you feed them can be used to train their global models, potentially exposing sensitive logs or IP addresses. Prévenance uses private, "Walled Garden" instances where your data never leaves your jurisdiction and is never used for training.
Strategic Expertise
AI excels at data processing but lacks contextual judgment. Understanding how M&A due diligence applies to your specific industry risk profile and your organisation, or whether a vendor control gap is genuinely material to your business, requires seasoned expertise. Our Human-in-the-Lead model ensures you get both speed and wisdom.
Prompt Engineering
We use complex, version-controlled prompts tested against "Gold Standard" datasets. A simple chat prompt cannot replicate the multi-layered analysis required for enterprise-grade financial analysis or M&A due diligence
Grounded Retrieval-Augmented Generation (RAG)
Public AI "hallucinates" because it tries to guess answers based on the entire internet. Our AI is "data-locked" via RAG. It is restricted to a private "Clean Room" containing only your documentation. If the answer isn't in your data, the AI will not invent a fact.
Source Tracing
Every single finding in Prévenance outputs includes a citation. This allows us to trace any output back to the exact log line, policy clause, or technical config that triggered it. You don't have to "trust" the AI; we can verify the source material instantly.
The Human Firewall
AI never has the final say at Prévenance. 100% of our deliverables undergo a rigorous review by Senior Subject Matter Experts who validate the logic, remove false positives, and ensure the advice is tactically sound.
Yes. We operate under a “Zero-Trust” data policy. All processing occurs in transient, locally hosted, isolated environments that are cryptographically wiped immediately after the engagement. We also provide an immutable audit trail—a “Chain of Reasoning”—for every interaction between our agents and your data.
Yes. We do not use international “Pass-through” APIs.
We adhere to a “Clean Room” protocol. Once the engagement is finalised and the report is signed off, the isolated environment containing your raw data is cryptographically shredded. We do not retain copies of your sensitive data for “future training” or analytics.
Prévenance does not operate in a vacuum. When you engage us, you are getting an experienced operator backed by our entire AI-native infrastructure and Human-in-the-Lead verification layer. They arrive with the analytical tools to diagnose issues and deliver results quickly and operate under our strict, board-ready reporting cadence.