# Principal Multi-Agent Security & Prompt Injection Defense Architect

- Link: https://accel-dna.com/jobs/principal-multi-agent-security-prompt-injection-defense/
- Published: 2026-10-29T15:43:00+00:00
- Author: Umar Mahmud

## Opportunity Overview

**Location:** 100% Remote (United States)
**Employment Type:** Full-Time Direct-
Hire via Accel DNA**Compensation Range:** $190,000 – $230,000 Base + Equity**Hiring
Agency:** Accel DNA – Premier Frontier Data & AI Talent Partner

Protect multi-agent autonomous enterprise deployments against indirect prompt injection,
data exfiltration, tool hijack, and automated adversarial threats.

## Key Responsibilities:

 * Design defense-in-depth architectural layers detecting and neutralizing indirect
   prompt injections in untrusted emails, PDFs, and web scrapes.
 * Implement least-privilege permission models and cryptographic verification for
   inter-agent communication channels.
 * Conduct automated adversarial red-teaming against agent decision trees, tool 
   invocation parameters, and memory stores.
 * Formulate real-time monitoring sensors flagging agent behavioral divergence and
   unauthorized privilege escalation.

## What You Bring:

 * 7+ years in application security, penetration testing, or software security, 
   with 2+ years specialized in LLM and agentic security.
 * In-depth knowledge of OWASP Top 10 for LLMs, jailbreak engineering, and prompt
   boundary hardening techniques.
 * Hands-on experience red-teaming autonomous systems and building programmatic 
   input sanitization filters.
 * BS/MS in Cybersecurity, Computer Science, or equivalent practical experience.

## How to Apply:

Interested in exploring this opportunity? Submit your profile or portfolio directly
to [contact@accel-dna.com](https://accel-dna.com/jobs/principal-multi-agent-security-prompt-injection-defense/contact@accel-dna.com)
or apply through [https://accel-dna.com/contact-us/](https://accel-dna.com/contact-us/).