Summary
What you’ll impact
The Security Engineering Manager owns and advances the security detection platform and the organization’s overall security posture. They use Python, security lakehouse architectures, SIEM-less data modeling, and cloud telemetry to develop detection logic, correlation rules, and benchmarks. Responsibilities include designing detection content, leading incident response, automating SOC 2 and ISO 27001 compliance evidence, and securing data ingestion pipelines. They build and mentor a security engineering team, collaborate with product and AI engineers, and set security standards organization‑wide.
Responsibilities
What you'll do
- You will own security at CipherData — the product's detection brain and the company's own security posture — reporting directly to the CEO, who is also our CISO.
- You define the detection content the agents reason over, the ground truth they are measured against, and the next generation of our security lakehouse and detection capabilities.
- You will work directly with the AI engineering team: they build the reasoning, you build what it reasons over and the bar it is held to.
- Around that core you also carry what a CISO's first engineer carries at a company our size: our own infrastructure and product security, customer-facing incident response, and the compliance work (SOC 2, ISO 27001, and customer-specific requirements) that our enterprise customers require.
- You will automate and delegate them, not staff a GRC team.
- Security lakehouse and detection platform: Drive the roadmap for our SIEM-less security lakehouse — schema and normalization across multi-vendor security data, retention and query strategy, and the detection and correlation layer that runs on it. Partner with product engineering on the data plane; own what "correct" means on top of it.
- Detection and correlation engineering: Build and maintain the detection logic and correlation rules that turn multi-vendor security data into signal our agents can reason over.
- Ground truth and evaluation: Curate and label real investigations into the benchmark datasets that tell us whether agent output is right — the standard the AI team builds against.
- Threat research into product: Track attacker tradecraft and convert it into agent capability, detection content, and test cases.
- Production SOC deployment: Deploy, tune, and validate AIDR in customer environments; own the loop that carries analyst feedback back into detection and evaluation.
- Customer incident response: Own the response to security incidents reported by customers or detected internally: analysis, root cause, remediation, customer communication, and post-incident review. Maintain the playbooks.
- Product and infrastructure security: Secure how we ingest, isolate, and handle customer security data — a privacy or tenancy failure is not something you recover from. Run assessments and pen tests on the product and our cloud footprint; integrate security into how we ship.
- Compliance and certifications: Own the technical side of SOC 2, ISO 27001, and customer security requirements, working with our Korea team and outside counsel. Automate evidence collection; don't build a paper program.
- Team: Build and grow the founding security team — hire, lead, and develop the security engineers who join you.
Requirements
What you’ll bring
- 8+ years in security engineering, detection engineering, threat hunting, or incident response, including senior-level SOC or detection-engineering work where you were accountable for outcomes
- Demonstrated detection engineering in production: you have written, tuned, and retired detections and can explain the false-positive trade-offs you made
- Deep hands-on experience with SIEM and security data platforms, EDR/XDR, and cloud and identity telemetry — including data modeling and normalization, not just query authoring
- Strong Python and comfort in an engineering codebase: code review, tests, CI
- Incident response leadership on real intrusions, including customer- or executive-facing communication
- Ability to articulate what separates a good investigation from a bad one precisely enough that it can be written down and measured
- Evidence of 0-to-1 ownership: you have built a function, program, or system from nothing, scoped your own work, and shipped without a large support structure
- Experience designing or operating a security data lake or lakehouse at scale
- Detection-as-code practice — version control, testing, and CI for detection content
- Experience building labeled datasets, benchmarks, or evaluations for security tooling or ML systems
- Familiarity with LLM-based systems and their failure modes (hallucination, prompt injection, fabricated evidence)
- Multi-tenant or MSSP environments; cloud security depth across major cloud providers
- SOC 2 / ISO 27001 experience from the engineering side
- Purple team, adversary emulation, or offensive security background
- Prior experience as a first or early security hire