AI & Security Platform ArchitectCompany: SecNinjaz

Role: AI & Security Platform Architect

About the roleSecNinjaz is building a cybersecurity agent platform that combines reliable tool execution, persistent security context, and independent evaluation with human review.

We are looking for a hands-on architect who can turn this direction into a practical, secure, and extensible product. You will define how agents execute work, use memory, integrate with security tools, and demonstrate trustworthy results. You will guide technical decisions while staying close to implementation, experimentation, and customer needs.

This is a platform-building role—not solely a web application penetration-testing role. Web and API assessment expertise is valuable, but the core requirement is the ability to design reusable capabilities for cybersecurity workflows.

What you will own

- Platform architecture: Define system components, interfaces, execution flows, and boundaries across the agent harness, tools, memory, evaluations, and backend services.
- Reliable agent execution: Design for state management, retries, recovery, long-running tasks, human approvals, and controlled tool use.
- Memory and context: Establish how security knowledge and application context are captured, retrieved, updated, isolated, and retired, with attention to provenance, freshness, and access control.
- Independent evaluation: Define how agent outcomes are assessed beyond an agent’s own claims, using reproducible tests, evidence, regression checks, and human review.
- Model and deployment qualification: Evaluate compatible models and deployment options against capability, security, latency, cost, and operational requirements. Keep the platform adaptable rather than unnecessarily tied to one provider.
- Security architecture:



Set appropriate controls for credentials, sensitive data, tenant boundaries, sandboxing, auditability, and authorized execution.
- Technical sequencing: Translate product goals into testable milestones. Separate capabilities needed for delivery from research that can proceed in parallel.
- Technical leadership: Review designs and code, build reference implementations, mentor engineers, and document significant decisions and trade-offs.
- Product alignment: Work with product and security colleagues to connect architectural choices to customer problems, measurable benefits, operating costs, and a credible path to scale.

What we are looking for
- Demonstrated experience designing and delivering production backend, distributed, or developer-platform systems.
- Hands-on experience building AI applications or agent systems that interact with tools, external services, and persistent state.
- Strong software engineering skills in Python and/or another backend language, with the ability to contribute production code.
- Sound judgment about where agentic execution is useful and where deterministic software or human review is more appropriate.
- Experience with retrieval, data architecture, or knowledge systems, including permissions, data lifecycle, and quality considerations.
- Ability to design meaningful evaluations, investigate failures, and use evidence to guide technical decisions.
- Strong security fundamentals, including threat modeling,



least privilege, secure tool integration, and handling untrusted inputs.
- Familiarity with deployment, observability, incident diagnosis, and performance and cost management.
- Clear communication: you can explain trade-offs, challenge unsupported assumptions, and work across research, engineering, security, and product teams.

Useful additional experience
- Cybersecurity tooling, penetration testing, vulnerability management, or security operations.
- Defending AI systems against prompt injection, unsafe tool execution, and sensitive-data exposure.
- Model serving, inference optimization, or comparative model benchmarking.
- Applied AI research that has been translated into working product capabilities.
- Building multi-tenant platforms or systems with strong isolation and audit requirements.

What success looks like
- A coherent architecture that supports working product delivery rather than remaining a design document.
- Reusable execution, memory, and evaluation capabilities with clear interfaces and security boundaries.
- Technical choices supported by practical testing and documented trade-offs.
- A platform whose reliability, quality, and operating costs can be measured and improved.
- An engineering team that can extend the product without repeatedly rebuilding its foundations.

How we assess candidatesOur process emphasizes practical architecture work, implementation depth, and security judgment. Candidates should be prepared to discuss systems they have built, explain failures and trade-offs, and work through a realistic design or engineering exercise.

Relevant certifications, publications, and open-source contributions are welcome, but none is mandatory or a substitute for demonstrated capability.

📌 AI & Security Architect (Delhi)
🏢 SecNinjaz Technologies
📍 Delhi