Application Development AI Agentic

# Senior Full-Stack AI Security Engineer / Technical Lead — AI Agent Cybersecurity Platform **Project:** The Grid by VetoGrid **Project Type:** Long-term contract with potential for an ongoing technical leadership role **Experience Level:** Expert **Engagement:** Milestone-based development, with potential for continued collaboration **Location:** Remote ## About VetoGrid VetoGrid is an emerging cybersecurity company focused on securing autonomous artificial intelligence and promoting responsible, accountable AI adoption. Our mission is simple: **Cybersecurity for the People. By the People.** We are developing **The Grid**, an AI security, governance, and accountability platform designed to help organizations discover, monitor, secure, and govern autonomous AI agents across multiple AI models and enterprise environments. The platform also aims to help organizations evaluate and manage the human impact of AI-driven decisions, including fairness and accountability in automated hiring systems. We have already started developing the application's frontend and core functionality using OpenAI Codex. We are now seeking an experienced technical professional to help advance the platform from its existing foundation into a secure, scalable, enterprise-ready product. ## The Opportunity We are looking for a highly skilled engineer with experience in cybersecurity, full-stack application development, AI/LLM integrations, and cloud architecture. The ideal candidate understands the security challenges created by autonomous AI agents and has experience building production-grade applications, API integrations, authorization systems, and security monitoring solutions. This is not simply a frontend development project. We need someone capable of designing and implementing the underlying technology that allows The Grid to function as a cybersecurity platform. ## Primary Responsibilities ### 1. Full-Stack Development - Review and improve our existing application and codebase. - Develop secure, scalable frontend and backend services. - Implement authentication, role-based access controls, and multi-tenant architecture. - Develop APIs, databases, administration features, and operational dashboards. - Improve application performance, maintainability, and reliability. ### 2. AI Agent Discovery & Inventory Develop capabilities to discover, identify, and inventory AI agents across enterprise environments. The platform should help organizations understand: - Where agents are deployed or hosted. - Which AI models and frameworks power them. - Who created, owns, and uses each agent. - Which applications, APIs, data sources, and systems they can access. - Their assigned business purposes, permissions, and risk classifications. - Whether an agent is authorized, unmanaged, or potentially operating outside approved governance. ### 3. Autonomous Agent Security & Threat Detection Help design and develop security capabilities that identify suspicious or unauthorized AI-agent behavior. Areas of focus include: - Prompt injection and agent manipulation. - Unauthorized tool execution. - Agent identity and privilege abuse. - Sensitive data exposure and exfiltration. - Agent-to-agent communication and delegated actions. - Behavioral anomaly detection. - Rogue or compromised agent identification. - AI-assisted threat investigation and correlation. ### 4. AI Security Enforcement Engine Help develop our proposed **Grid Autonomy Firewall**. The objective is to evaluate agent requests against independently enforced security policies before potentially harmful actions occur. Capabilities should eventually include: - Policy-based authorization. - Agent identity verification. - Tool and API access restrictions. - Human approval workflows. - Action blocking at supported enforcement points. - Containment and credential revocation through authorized integrations. - Comprehensive security decision logging. ### 5. Enterprise Integrations Develop a modular integration architecture capable of supporting: - OpenAI, Anthropic Claude, Google Gemini, and other AI providers. - Microsoft Azure and AWS AI services. - Custom and self-hosted AI agents. - Model Context Protocol (MCP). - Agent-to-Agent (A2A) communications. - Enterprise identity systems such as Microsoft Entra ID. - Cloud resources, databases, APIs, and SaaS platforms. We understand that integration depth and enforcement support vary by provider. The architecture should accurately represent those limitations. ### 6. AI Governance, Fairness & Accountability We are also developing a human-centered AI accountability capability. This includes creating technical foundations for organizations to evaluate their own AI systems for responsible use. An initial focus is **employer-side AI hiring systems**, including: - Automated applicant screening. - Candidate ranking algorithms. - AI-assisted interview assessments. - Potential discriminatory outcomes. - Fairness and bias testing. - Human oversight and appeal workflows. - Audit evidence and regulatory governance. We are not developing tools to evaluate or penalize job applicants for using AI to improve their résumés. Our focus is accountability for organizations building and deploying consequential AI decision systems. Experience with Responsible AI, algorithmic auditing, or AI governance is highly desirable. ## Preferred Technical Experience - **Frontend:** React, Next.js, TypeScript, Tailwind CSS. - **Backend:** Python, FastAPI, Node.js, or Go. - **Databases:** PostgreSQL and secure multi-tenant database design. - **AI/LLMs:** OpenAI APIs, Anthropic APIs, LLM orchestration, agent frameworks. - **AI Protocols:** MCP, A2A, tool calling, and agent execution frameworks. - **Security:** OAuth 2.0, OIDC, IAM, Zero Trust, authorization, API security. - **Policy Engines:** Open Policy Agent or similar policy-as-code solutions. - **Cloud:** AWS, Azure, Docker, CI/CD, infrastructure as code. - **Monitoring:** OpenTelemetry, structured event logging, security analytics. - **Development:** GitHub, automated testing, secure SDLC, and AI-assisted development tools such as Codex. ## Ideal Candidate We are particularly interested in someone who: - Has built a SaaS platform or cybersecurity product beyond the prototype stage. - Understands enterprise security architecture and threat modeling. - Has hands-on experience building applications involving autonomous agents or LLMs. - Can explain complex engineering decisions clearly. - Can independently evaluate existing code and recommend improvements. - Prioritizes security, privacy, reliability, and software quality. - Can work collaboratively with a cybersecurity-focused founder. - Is comfortable with a phased, milestone-based development approach. ## Initial Deliverables The first engagement will focus on establishing a reliable technical foundation and a development roadmap. **Milestone 1 — Technical Assessment** - Review the existing source code, infrastructure, authentication, and database. - Identify security weaknesses, architectural gaps, and technical debt. - Recommend an implementation roadmap and effort estimates. **Milestone 2 — Core Platform Improvements** - Improve application reliability, security, and backend functionality. - Refine the agent registry, ownership workflows, policy management, and governance modules. - Add or improve automated tests and deployment processes. **Milestone 3 — AI Agent Integration** - Implement initial agent and identity integrations. - Collect verified agent inventory and telemetry. - Demonstrate agent discovery and ownership correlation. **Milestone 4 — Security Enforcement Proof of Concept** - Develop a testable policy enforcement gateway. - Demonstrate interception and blocking of unauthorized agent actions in a controlled environment. - Integrate policy decisions and evidence into The Grid's dashboard. **Future Milestones** - AI behavioral threat detection. - Multi-agent attack correlation. - Enterprise deployment capabilities. - AI fairness and accountability assessments. - Expanded cloud and AI provider integrations. ## Project Expectations All development must follow secure coding practices, use version control, include appropriate documentation, and undergo testing. The source code and intellectual property developed under the engagement must be owned by VetoGrid, subject to agreed contractual terms. We expect clear technical communication, transparent progress reporting, and milestone demonstrations. ## To Apply Please include: 1. Examples of SaaS, cybersecurity, or AI-agent platforms you have personally developed. 2. Your experience with LLM security, agent frameworks, MCP, or runtime authorization. 3. Your experience building secure enterprise applications. 4. Your proposed approach to reviewing and extending an existing application. 5. Your availability, hourly rate, and estimated effort for the initial technical assessment. **Important:** Please describe your direct engineering contributions to the projects you reference. We are looking for someone interested in helping develop a meaningful cybersecurity product—not simply completing a short-term website development assignment. **VetoGrid — Cybersecurity for the People. By the People.**