Job Title : Sr Solutions Architect
Location: Irving TX (3 days onsite – Hybrid role)
Duration 12 months
Must Have skills
The person has to understand post-quantum cryptography well enough to judge what a remediation means. That means NIST FIPS 203 (ML-KEM), 204 (ML-DSA) and 205 (SLH-DSA), the NIST IR 8547 transition timeline, and the hybrid key-exchange policies that will apply during migration. The architect must also know TLS 1.3, AES-256, and how to read a CBOM (cryptographic bill of materials), since each finding is the gap between an application’s CBOM and a master target baseline. Agentic AI: the architect designs a multi-agent LangGraph workflow in which scan, plan, remediation, human approval, execution and re-scan agents work through each finding. The agents fill in versioned remediation templates rather than inventing fixes, and guardrails are enforced by the pipeline, not by prompts. The LLM layer runs on Amazon Bedrock, with retrieval over OpenSearch and state held in DynamoDB. AWS: the solution runs in an InfoVision AWS sandbox on ECS Fargate, with EKS hosting the SOE portal replica, plus API Gateway, Lambda and Cognito for the reviewer dashboard. The architect must design the networking, VPC endpoints, load balancers, KMS and Secrets Manager, and the Terraform and CodeBuild execution path, all while keeping Client’s raw data inside Client’s boundary.
About the Role:
We are looking for a Senior Solution Architect to design an end-to-end Agentic AI pipeline for an enterprise cybersecurity POC. You will be the technical authority responsible for architecting a multi-agent workflow that detects, plans, and remediates cryptographic vulnerabilities using Large Language Models and AWS infrastructure.
Key Responsibilities:
Architect a multi-agent LangGraph workflow hosted on AWS ECS Fargate.
Design the LLM integration layer utilizing Amazon Bedrock (Claude/Gemini APIs) for automated vulnerability remediation logic.
Architect a Retrieval-Augmented Generation (RAG) vector system using OpenSearch to map security scan findings against enterprise rule dictionaries.
Design a robust state-management layer using DynamoDB to track remediation checkpoints.
Design a secure "Reviewer Dashboard" using API Gateway, AWS Lambda, and Amazon Cognito for human-in-the-loop approval gates.
