Sr. AI Security Engineer
Backblaze External WebsiteResponsibilities
- Architect and implement guardrails for tool-using AI systems, including tool access controls, allowlists, context and memory isolation, and step-level action validation.
- Build runtime enforcement mechanisms such as interceptors, proxies, middleware, policy decision and enforcement layers, rate limits, execution bounds, and kill-switches.
- Design identity and access controls for agents and automation using short-lived credentials, scoped permissions, human/non-human access separation, and task-bound identity.
- Make AI actions attributable and auditable through logging and tracing of prompts, tool usage, and decision flows.
- Build behavioral baselines, anomaly detection, and alerts for abnormal tool usage, suspicious prompt patterns, and unexpected data access.
- Threat-model agentic systems with MAESTRO by mapping capabilities, trust boundaries, attack paths, misuse cases, and adversarial scenarios.
- Translate threat-model findings into safeguards and detection logic.
- Protect developers from sensitive data exposure and unsafe AI-generated code or automation, and enable controlled use of Claude Code, Codex, and Cursor.
Requirements
- 7+ years of experience in security engineering or backend systems.
- Proven experience designing and deploying security controls, including runtime enforcement layers and identity and access systems for non-human entities.
- Strong programming skills; Python is preferred, with Go, Java, or TypeScript as additional options.
- Experience using AI-assisted development tools such as Claude Code in real workflows and understanding their security risks and safeguards.
- Experience with logging, monitoring, detection systems, and securing API or service interactions.
- Practical familiarity with agentic AI systems or tool-integrated LLM workflows and OWASP guidance for AI and agent risks.
- Applied experience with OWASP Agentic AI or LLM risk guidance, NIST AI RMF concepts, and CSA guidance on workload and machine identity.
- Strong understanding of Zero Trust for non-human identities, secrets management, credential scoping, and observability tooling such as OpenTelemetry and ELK.
- Bonus: experience securing internal AI platforms or developer-facing AI tools, detection engineering, threat hunting, adversarial testing, agent frameworks such as LangChain or LlamaIndex, or mentoring engineers.
Benefits
- The role is identified as LATAM.
- Backblaze is an equal opportunity employer committed to diversity, equity, and inclusion.
Tech Stack
Categories
About Backblaze External Website
AI infrastructure runs on data. Storage gets hard at scale. We’ve been breaking, rebuilding, and improving that storage layer for 20 years. We started by building storage servers from scratch in a Palo Alto apartment to save 90% on hardware. Turns out hard constraints make for pretty good engineering. That discipline forged into a platform that performs as well or better than competitors, scales to multi-exabyte workloads, and costs a fraction of what AWS or Google charge. No blank checks or architectural sprawl. Today, the world's leading AI infrastructure providers, neoclouds, and data-intensive businesses run on Backblaze. Our platform includes: - B2 Overdrive: High-throughput object storage engineered for AI infrastructure, GPU workloads, and massive-scale data pipelines. - B2 Neo: White-label enterprise cloud storage powering neocloud platforms for AI, HPC, and media. - B2 Cloud Storage: S3-compatible object storage with predictable pricing, free egress, and no lock-in. - Computer Backup: Automatic, continuous endpoint backup for Macs and PCs. 500,000+ customers. 175 countries. Nasdaq: BLZE. backblaze.com