With 4Minds, you can advance beyond AI trained on public knowledge to custom AI that actually understands how your unique business operates. The platform continually applies evolving enterprise knowledge and operational context, reasoning across relationships, dependencies and changing business conditions. With 4Minds, you can deploy AI that stays aligned with missioncritical operations, supports strategic decisions and modernizes business workflows while continuously improving over time.
About This Role
We're seeking a DevSecOps Engineer to secure the full lifecycle of our AI platform, from the data that feeds our models to the fine-tuned models running inside customer environments. This is not a typical SaaS security role. Our product trains, adapts, and ships models that learn continuously from live enterprise data, and it runs on-prem, in customer clouds, and in restricted or air-gapped networks. That means the attack surface includes training data, model weights, knowledge graphs, inference endpoints, and the pipelines that connect them.You'll build security into how we ingest data, train and fine-tune models, package and distribute them, and operate them in production. You'll work closely with ML engineers, platform engineers, and Forward Deployed Engineers, and you'll help customers' security teams trust what we deploy into their environments.This role sits between mid to senior level. You'll own well-defined parts of our security program end to end, with guidance from senior engineering leadership, and grow into broader ownership as our AI security practice matures.
Key Responsibilities
- Secure connectors and pipelines that pull customer data from sources such as Microsoft, Databricks, AWS, and Google
- Enforce encryption, access control, and tenant isolation for training data, embeddings, and knowledge graph stores
- Build controls that detect data poisoning, malicious inputs, and sensitive data (PII/PHI) before it reaches training
- Track data lineage and provenance so every model can be traced back to the data that shaped it
- Harden GPU training infrastructure, job schedulers, and notebook environments
- Manage secrets, credentials, and least-privilege access for training and fine-tuning workloads
- Secure the continuous-learning loop so live data updates cannot silently degrade or compromise models
- Scan and verify third-party models, datasets, and ML libraries before use (e.g., unsafe serialization formats, malicious weights, vulnerable dependencies)
- Sign and attest model artifacts and containers; maintain SBOMs and model bills of materials (ML-BOMs)
- Operate secure model and artifact registries with versioning and integrity checks
- Embed SAST, SCA, secrets scanning, IaC scanning, and container scanning into CI/CD
- Add AI-specific security gates: prompt injection and jailbreak tests, data leakage checks, and model evaluation before promotion
- Support adversarial testing and red-teaming of models and agentic features
- Package the platform securely for on-prem, customer-cloud, and air-gapped deployments (Kubernetes, Helm, Terraform)
- Harden inference endpoints, API gateways, and model-serving infrastructure
- Partner with Forward Deployed Engineers to answer customer security questionnaires and pass customer security reviews
- Monitor for model abuse, anomalous inference patterns, data exfiltration, and model drift with security impact
- Contribute to incident response runbooks covering AI-specific incidents such as poisoned data, leaked weights, or prompt injection exploits
- Manage vulnerability remediation across infrastructure, containers, and ML dependencies
- Implement and evidence controls for SOC 2 Type II, ISO 27001, ISO 42001, GDPR, and CCPA
- Map controls to AI risk frameworks such as NIST AI RMF, OWASP Top 10 for LLM Applications, and MITRE ATLAS
- Automate compliance evidence collection and policy-as-code wherever possible
Required Qualifications
- BS in Computer Science, Cybersecurity, Engineering, or related field, or equivalent practical experience
- 4–6 years of experience in DevOps, platform, cloud, or security engineering, with at least 2 years focused on security
- Hands-on experience building and securing CI/CD pipelines (GitHub Actions, GitLab CI, or similar)
- Working knowledge of Kubernetes, containers, and infrastructure as code (Terraform, Helm)
- Experience with at least one major cloud (AWS, Azure, or GCP), including IAM, networking, and key management
- Practical use of security tooling: SAST, SCA, secrets scanning, container and IaC scanning
- Scripting or programming ability in Python plus Bash or Go
- Understanding of how ML systems are built: training data, model artifacts, model serving, and inference APIs
- Familiarity with AI-specific threats such as data poisoning, prompt injection, model theft, and sensitive data leakage
- Clear written communication; able to explain risk to engineers and to customer security teams
- Ownership mindset and comfort working in a fast-paced startup
Preferred Qualifications
- Experience securing MLOps tooling such as MLflow, Kubeflow, Ray, Databricks, or SageMaker
- Experience with GPU infrastructure and model-serving stacks (e.g., vLLM, Triton, KServe)
- Exposure to LLM security testing, red-teaming, or guardrail tools
- Experience with artifact signing and provenance (Sigstore/Cosign, SLSA, in-toto)
- Experience shipping software into on-prem, customer-managed, or air-gapped environments
- Hands-on work supporting SOC 2, ISO 27001, or ISO 42001 audits
- Familiarity with knowledge graphs, vector databases, or RAG architectures
- Experience with policy-as-code (OPA/Gatekeeper, Kyverno) and runtime security (Falco or similar)
- Security certifications such as CKS, AWS Security Specialty, or Security+
- Prior startup experience or experience at an AI/ML company
Compensation
$130,000 – $220,000 per year
- Base salary as listed per role, plus equity package
- Performance-based commission structure for quota-carrying positions
- Annual merit-based performance reviews targeting 20% of base salary for non-quota-carrying roles
- Comprehensive healthcare coverage (80% employer-paid)
- 401(k) with company matching
- Flexible PTO starting at 15 days
- 11 paid holidays
Professional Development
- Annual training and certification budget
- Access to enterprise learning platforms
- Conference attendance opportunities
- Internal workshops and knowledge-sharing sessions
- On-site at our Dallas, TX office, 5 days per week
- Standing desks and modern collaborative workspace
- Monthly team events
- Company-provided laptop
Interview Process
- Technical interview loop (half or full day)