// Discipline 04
Security
Cybersecurity and hardening
Infrastructure designed to fail well: controlled attack to find the flaws before others do, and defense designed to keep working when something gives way.
01 · Who it is for
Who it is for
Companies with critical infrastructure
That cannot afford outages or data leaks.
Development and platform teams
That want security by design, not a patch at the end.
Fintechs and startups
That need to pass an audit or protect customer data.
02 · How we work
How we work
Pentesting and red teaming
Controlled, authorized attack on applications, networks and cloud.
Hardening
Secure configuration of servers, networks and cloud services.
Resilient architecture
Design of fault-tolerant systems and recovery plans.
Audit and training
Code and configuration review, security of LLM applications and workshops for technical teams.
03 · Specialties
Specialties
- Pentesting and red teaming
- Server and network hardening
- Cloud security
- Mission-critical software
- Fault-tolerant architectures
04 · Method
Method
- 01
Diagnosis
What data exists, which decision depends on it and what has failed so far.
- 02
Design
Hypothesis, method and success criteria in writing, before building anything.
- 03
Build and test
Reproducible code and analysis, tested against adverse cases.
- 04
Handover
Everything is documented and in your team’s hands.
05 · Sectors
Sectors
Finance
- Protecting trading platforms and customer data
- Audit, DORA and NIS2 readiness (advisory)
Defense
- Hardening of sensitive infrastructure
- Review of mission-critical software
Civil
- Cloud security for companies and startups
- Service continuity and failure recovery
06 · Use cases
Use cases
Illustrative cases: they describe the kind of work we do, not specific clients.
Find the attack paths
- Problem
- Accumulated permissions nobody reviews.
- Approach
- Graph of identities and permissions, paths to critical assets and a controlled test.
- Deliverable
- Prioritized list of paths and how to cut them.
Harden a platform
- Problem
- Servers and services running default configurations.
- Approach
- Review against reference guides, tested and documented changes.
- Deliverable
- Hardened platform and a maintenance guide.
Design to fail well
- Problem
- A single failure that takes down the whole service.
- Approach
- Single points of failure, redundancy and controlled chaos tests.
- Deliverable
- Reviewed architecture and a tested recovery plan.
Penetration test
- Problem
- Knowing what an attacker would see before they see it.
- Approach
- Authorized pentest with agreed scope and rules.
- Deliverable
- Technical and executive report with fix priorities.
07 · Data and topology
Data and topology
A network is a graph of machines, identities and permissions. Attack paths are paths in that graph: we find them and cut them where it pays off most.
08 · Research
Research
No publications yet. The first one is in preparation.
A data problem in this area?
Tell us the context: what data exists, which decision depends on it and what has failed so far.