Want to know if this job is worth applying to?
Paris, Île-de-France, France
Hybrid
Reporting to the CISO, you'll own the security roadmap for Pigment's product, infrastructure, and CI/CD environment, and you'll build much of it yourself.
This is a hands-on role: expect to read code, threat model new services, reproduce and triage vulnerabilities, dig through infrastructure configuration and build the automation that closes gaps.
You'll also set direction and bring product and engineering with you, but that influence comes from technical credibility and not process: the engineers you work with will take you seriously because you've been in the same code they have.
The scope is broad: application security, infrastructure security, detection and response, and the assurance work that keeps our certifications standing. You won't be covering it alone. The security team is seven people and growing, with colleagues already owning compliance, governance and security operations, so this role can go deep on product and infrastructure and not be spread too thin across everything. Nobody arrives fluent in all of it: we're looking for real depth in several of these areas and the judgement to grow into the rest.
Key responsibilities include:
Product & Infrastructure Security Design - Design security features into the product itself and strengthen defence-in-depth across the platform. Threat model new services before they ship, and make the architectural calls that are expensive to reverse later.
Security Review & Risk Assessment - Review code, architecture and configuration yourself, and be the person developers and PMs bring problems to early rather than late. Deliver solutions that balance risk against business benefit, and escalate the calls that genuinely need senior arbitration.
Assurance & Testing - Run our assurance programme: internal code, architecture and configuration reviews, red team exercises, and the bug bounty. Own the relationship with third-party auditors, and measure the control KPIs that keep our certifications defensible.
Vulnerability Management - Own vulnerabilities from detection through to verified fix: reproduce them, score and triage them, design or validate the mitigation, and confirm it actually worked. Improve the KPIs that tell you whether the process is holding.
Detection Engineering - Build and improve our detection capability alongside the infrastructure and engineering teams: identify the signals worth collecting, write rules that catch real attacks without drowning us in noise, and build the response playbooks behind them.
Incident Response - Lead security investigations into the production environment end to end, from first signal to root cause, across incidents and fraud. Flag the repetitive work worth automating and the detections worth building, and hand those to the security operations team.
SDLC - Set the technical direction for how engineers at Pigment build securely: guidance, paved paths, and reviews that teach and not just gate. Contribute to company-wide awareness where it counts, but your primary audience is product, engineering and SRE.
Example projects that would fall under your responsibilities (actual examples of recently completed or currently on the roadmap):
Secure the design and development of our AI features, including the MCP Server and Modeler Agent: threat modelling, design reviews, working alongside the engineers building them, and security assessments.
Work out where AI-assisted analysis actually belongs in our pipeline. AI-powered security reviews reason about a diff semantically rather than pattern-matching it, which is genuinely different from classic SAST - but it's not a drop-in replacement, and figuring out the split between AI review, traditional SAST and SCA is an open question we'd like you to answer with evidence rather than vendor claims.
Migrate GitHub to managed identities: provisioning through Okta, retiring personal accounts and long-lived PATs, and moving CI to short-lived OIDC credentials instead of stored keys.
Design agent identity for the MCP Server - delegated access tokens, token exchange, and making sure an agent acting for a user can never exceed what that user could do themselves.
Push least privilege further across production and CI/CD. Better than it was, not where we want it.
For a concrete example: here's how we built a sandboxed execution environment for LLM-generated code.
Technical Environment
Mostly production, with the occasional internal IT-adjacent project:
Main sites in Paris, London, Toronto and NYC
MacOS, Windows, Linux workstations
GCP, Kubernetes, Terraform, Postgres, SingleStore, Vault
Okta, OAuth, JWT, C#, .NET Core, TypeScript, React, Python, Go
Datadog (SIEM), CloudFlare ZTNA, Falco, Wiz, Riot, HackerOne, TruffleHog
Google Workspace, Jumpcloud, Vanta, Hibob, Slack
GitHub, CircleCI, ArgoCD
SOC 1, SOC 2, ISO 27001, ISO 27017, ISO 27018, ISO 42001
Who you are
You have at least 8 years of experience in security as a Security Engineer, Pentester or Security consultant, with real depth in product and/or infrastructure security. We care more about the scope of what you've owned than the exact number of years.
What you've done
Owned a security roadmap for a product or platform end to end, rather than delivering items on someone else's
Driven security work through engineering teams you had no authority over
Been the person your organisation escalated to on hard architectural calls
Worked hands-on across a broad technical surface - development, databases, networking, web
How you work
You're hands-on and intend to stay that way (this position does not include people management)
You look for the workable answer, not the blocking one
You have a good dose of humility, and you help the people around you get better
You speak English fluently, French is a strong plus
How we hire
After a first call with our recruiter, five conversations, around four hours in total, usually over two to three weeks:
Ways of working (45 min) — how you operate with people, and what draws you to this role specifically.
Technical deep dive (1h30) with a security engineer — including a practical. We'll give you a real problem we've already solved and talk through how you'd approach it.
Cross-functional (45 min) with an engineer from product or platform — someone who'd be on the receiving end of your work.
Ownership and career (1h) with the CISO and a member of the security team.
Final (45 min) with our co-founder and CTO.