AngeloDe LeonSoftware
& applied AI
Cybersecurity consulting · San Francisco

A stronger security posture. A clearer plan.

I help businesses understand their exposure, fix practical security gaps, and make safer decisions about websites, accounts, infrastructure, and AI. The work starts with a scoped review and produces actions your team can understand and maintain.

Scope and authorization first
Prioritize practical fixes
Document changes and ownership
San Francisco · remote collaboration
What I do

Six areas to review before a small gap becomes a problem.

Security work should account for your actual systems and the consequences of a change. I focus on clear findings, prioritized remediation, and a useful handoff.

Security posture reviews

Review the assets, accounts, configurations, and business workflows in scope. Document what was checked, the evidence behind each finding, and practical remediation priorities.

A prioritized action plan

Identity & access hardening

Review multi-factor authentication, administrative roles, account recovery, shared access, and offboarding. Reduce unnecessary access and document who owns each system.

Clearer control of important accounts

Website & application security

Assess agreed authentication flows, exposed files, dependencies, deployment configuration, and application behavior. Review AI-generated code, implement practical hardening, and verify fixes. Penetration testing uses defined targets, methods, and explicit authorization.

Verified application security improvements

Server testing & network hardening

Perform authorized server penetration testing and review VPS, container, firewall, patching, and remote-access configuration. Harden local AI hosts and configure Tailscale or comparable private network access with defined users and permissions.

Tested exposure and clearer access boundaries

Domain, email & recovery readiness

Review DNS ownership, email authentication, backups, restore procedures, and recovery access. Check that the team can find the documentation and knows what to do when something fails.

Better prepared for disruption

AI penetration & jailbreak testing

Assess authorized models and custom AI applications for jailbreaks, prompt injection, unintended disclosure, and tool-access failures. Define test boundaries, preserve reproducible evidence, recommend mitigations, and retest agreed fixes.

Scoped findings for AI-specific risks
In practice

Security decisions in the context of real work.

Explore a typical workflow. These are illustrative scenarios; the details and results depend on the agreed project.

Growing teams

Before

Important accounts have shared logins, unclear owners, and inconsistent offboarding.

A possible next step

A reviewed access model establishes named owners, appropriate permissions, MFA, and a documented process for joining or leaving the team.

What it involves
  • Access review
  • MFA
  • Recovery documentation

Websites & applications

Before

An application is live, but exposed files, dependencies, and configuration have not had a focused review.

A possible next step

The agreed scope is assessed, findings are prioritized by impact and evidence, and fixes are tested before a controlled release.

What it involves
  • Configuration review
  • Dependency review
  • Verification

Cloud infrastructure

Before

A server or container stack grew over time, leaving services, credentials, and maintenance responsibilities difficult to track.

A possible next step

A scoped inventory and hardening plan clarifies exposure, patching, permissions, backups, and how the team recovers access.

What it involves
  • Docker & VPS review
  • Firewall configuration
  • Backup planning

AI adoption

Before

Staff use several AI tools without a clear policy for data, connected accounts, or approval of actions.

A possible next step

Approved tools and data boundaries are documented, access is scoped, and people practice when to review or escalate an AI-generated action.

What it involves
  • AI risk review
  • Scoped integrations
  • Team workshops
How it works

Review carefully. Fix deliberately. Verify the result.

01

Agree the scope

Define the systems, authorization, access, constraints, and business priorities before assessment work begins.

02

Assess & prioritize

Collect evidence, distinguish confirmed findings from hypotheses, and rank improvements by risk and effort.

03

Remediate & verify

Make agreed changes through staging or a controlled change process, then retest the affected behavior.

04

Document & hand off

Record the changes, remaining issues, ownership, recovery steps, and any follow-up support scope.

Evidence and scope

A review should leave you with a usable plan.

Security findings need context. I explain what was observed, what remains uncertain, and what a change will affect before recommending the next step.

  • Work only on explicitly authorized systems and agreed tests
  • Keep credentials and client evidence out of public repositories
  • Plan backups, rollback, and recovery around important changes
  • Define specialist, incident response, or compliance work separately when needed
security-review.yml · illustrative example
scope: approved systems and test methods
evidence: findings with reproducible context
priority: business impact + exposure
changes: agreed, reversible where possible
verification: retest affected behavior
handoff: owners, recovery, remaining risks
Ways to start

Pick the step that fits.

Start with a focused review, a defined project, or ongoing support. The scope, costs, and responsibilities are agreed before work begins.

Understand the exposure

Focused security review

Fixed scopeSystems and assessment boundaries agreed first

Get an evidence-based view of a website, deployment, account environment, or AI workflow.

  • Scope and access planning
  • Findings with context
  • Prioritized remediation plan
Scope a security review
Maintain the baseline

Ongoing security support

Ongoing engagementResponsibilities and response expectations defined

Review changes and maintain a practical security baseline as your systems evolve.

  • Periodic configuration reviews
  • Access and patching priorities
  • AI and workflow risk reviews
  • Staff guidance and documentation
Discuss ongoing support
Questions

A few useful answers.

What cybersecurity services do you provide?

I offer scoped security posture reviews, identity and access hardening, website and application reviews, cloud and server configuration work, recovery readiness, and AI risk reviews. Each engagement defines the systems, authorized methods, deliverables, and verification work in writing.

Do you offer penetration testing or incident response?

I offer authorized penetration testing for agreed server, application, and AI-system targets, including jailbreak assessments for models and custom AI applications. Scope, test methods, access, operating constraints, and reporting are agreed before work starts. Emergency incident response or a staffed SOC is a separate service requirement.

Can you guarantee that we will not be breached?

No provider can guarantee that. The objective is to understand exposure, reduce avoidable risk, and improve the ability to detect, respond, and recover. A review records the work performed and its limitations rather than presenting a permanent security guarantee.

Can you review our existing website or AI-generated application?

Yes. I can assess agreed parts of an existing application, its dependencies, access flows, deployment, and connected tools. Findings are evaluated against the actual context, and remediation can be scoped and verified separately.

Will security changes disrupt our business?

Changes can affect access and integrations, so we plan them around the system and your operating requirements. Appropriate work includes backups, staging or controlled rollout, rollback notes, verification, and a clear owner for recovery access.

Can this help with compliance or cyber insurance requirements?

A review can help identify technical and operational gaps relevant to your requirements and produce useful evidence of remediation. Legal interpretations, formal certifications, insurer approval, and regulated assessments require the relevant qualified parties; they are not implied by a general security review.

Can hardening make our application compliant?

Hardening can address technical gaps against agreed industry or organizational requirements and produce evidence of the work performed. Compliance also depends on policies, operations, contracts, and the applicable assessment process. I coordinate technical readiness with your responsible advisers; I do not promise blanket compliance or issue regulatory certifications.

Let’s make a plan

Your next step.
Made clearer.

Tell me what you are working on, what is getting in the way, and what a useful result would look like.

Based in San Francisco. Available for Bay Area projects and remote collaboration.