DevSecOps Engineer

**The Role** Prodigal processes sensitive borrower conversations and financial data at scale — security here is a product requirement, not a compliance checkbox. You'll work alongside our DevSecOps Lead to embed security across our cloud infrastructure, CI/CD pipelines, Kubernetes workloads, and AI systems. That means owning IAM policies, hardening our AWS environment, driving SOC2 compliance, and building the guardrails the engineering team ships within. What makes this role genuinely different: you'll also operate real-time voice AI infrastructure at production scale — sub-1 second latency pipelines, GPU workloads, model serving, and inference cost management. That combination of security ownership and AI/ML infrastructure exposure is hard to find. It's a small team with direct access to senior engineering leadership. You'll have real ownership, not a ticket queue. **What You’ll Do** - **AWS Infrastructure:** Manage and optimise AWS resources — compute, networking, storage. Implement infrastructure-as-code via Terraform. Own cost monitoring, rightsizing, and cloud spend optimisation. - **Cloud Cost Management:** Drive overall cloud cost optimisation efforts, including identifying opportunities for rightsizing, work with internal teams to optimise usage of infra, eradicate waste, optimise workflows. - **Kubernetes & Container Orchestration:** Operate and troubleshoot Kubernetes clusters. Manage pod scheduling, resource limits, autoscaling (Karpenter), service mesh (Istio), and container lifecycle. - **Monitoring & Incident Response:** Build and maintain monitoring, alerting, and dashboards (Prometheus, Grafana). Participate in incident response, root cause analysis for infrastructure components, and runbook development. - **CI/CD & Deployments:** Architect and maintain automated pipelines with integrated security guardrails, including SAST, dependency scanning, and secrets detection. Orchestrate seamless deployments across staging and production environments using TrueFoundry. - **AI/ML Infrastructure Support:** Assist with DataBricks environment management, GPU workload scheduling, and model deployment pipelines. Help track and optimise AI compute costs across hosted and self-hosted workloads. - **Automation & Tooling:** Identify manual, repetitive operational tasks and automate them. Build internal tooling and scripts (Python, Bash) that make the engineering team more productive. **What We’re Looking For** - 3-5 years of hands-on DevOps/DevSecOps experience in a product company or startup. B.Tech, BE, or M.Tech in Computer Science or related discipline. - AWS proficiency — EC2, VPC, IAM, S3, CloudWatch. Comfortable in the console and CLI daily, not just during incidents. - Kubernetes in production — deployment management, troubleshooting, scaling. You've operated clusters, not just followed tutorials. - CI/CD pipelines — GitHub Actions, GitLab CI, ArgoCD, or similar and Embed automated security testing (SAST/DAST) into CI/CD workflows to catch vulnerabilities before they reach production. - Infrastructure-as-Code — Terraform or CloudFormation. You write and review IaC; you don't manage infra through the console. - Scripting — Python and/or Bash for automation and tooling. You reach for a script before doing something manually twice. - Monitoring & Observability — Prometheus, Grafana, or similar. You've set up dashboards, alerts, and runbooks from scratch. - Security fundamentals — working knowledge of IAM least-privilege, secrets management, and basic compliance concepts (SOC2, PCI-DSS). - Project ownership — you can take an ambiguous infra problem, break it into a plan, and drive it to completion without hand-holding. In a small team, that's not optional. - Communication & documentation — your runbooks, PRs, and architecture notes are readable by someone who wasn't in the room. **Bonus Points** - Hands-on with ML/AI infrastructure — GPU instances, model serving (TorchServe, Triton, TrueFoundry), or inference workload optimisation. - DataBricks or Spark experience — cluster management, job scheduling, cost controls. - SOC2 or PCI-DSS compliance experience — familiarity with controls, evidence collection, and audit processes. - Relevant certifications — AWS Solutions Architect, CKA, AWS Security Specialty, or CompTIA Security+. - You think about automating. **Why Prodigal** - YC + Accel backed, strong product-market fit in a massive lending vertical. - Real-time voice AI at production scale — sub-1s latency pipelines. Genuinely interesting infra, not routine cloud babysitting. - AI/ML infrastructure exposure from day one — GPU workloads, composite cost optimisation, model serving. Career-differentiating experience at the 3–5 year mark. - Two-person DevSecOps team: you’ll ship real infrastructure with direct impact, not disappear into a 20-person ops org. - Work directly with senior engineering leadership — short feedback loops, fast decisions. - Competitive compensation and ESOPs.