The Solution Architect – Agentic AI will drive the design, architecture, and implementation of an enterprise -grade AI agent ecosystem within Mercedes -Benz.
This role bridges architecture, engineering, AI/ML, cloud, DevSecOps , security, and business enablement, ensuring that AI agents are secure by design , scalable, interoperable, and aligned with enterprise and regulatory standards.
The architect collaborates closely with System Architects, product owners, developers, platform teams, security, and business stakeholders to deliver agentic solutions across multiple domains while enforcing shift -left security, automation, and continuous compliance.
Key Responsibilities1. Architecture & Solution Design
§ Architect end -to-end agentic AI solutions with security, compliance, and operability built in by default .
§ Create blueprints, reference architectures, and reusable patterns for secure, cross -cloud agent deployment .
§ Define data, model, tool, and API integration patterns with secure -by-design principles .
2. Multi -Cloud & Platform Expertise
§ Architect AI solutions across Azure and AWS , aligned with MB’s AI Ecosystem.
§ Ensure cloud -agnostic, DevSecOps -enabled deployments with consistent security controls and observability across environments.
3. Technical Leadership
§ Guide engineering teams in implementing agent capabilities, APIs, pipelines, and integrations following DevSecOps best practices .
§ Review solution designs for quality, scalability, security posture, cost efficiency, and enterprise alignment .
4. DevSecOps, Governance, Security & Compliance (Primary Responsibility)
§ Design and enforce DevSecOps pipelines for AI/ML and agentic systems, integrating security, quality, and compliance checks into CI/CD.
§ Implement enterprise standards for: o Secure SDLC, IaC , and pipeline governance o IAM/RBAC, secrets management, encryption, audit logging o Data privacy, model governance, and EU AI Act alignment
§ Drive shift -left security , threat modeling, vulnerability management, and continuous compliance for agentic systems.
5. Cross -Functional Collaboration
§ Work closely with System Architects, Product teams, Platform teams, Security, and external partners.
§ Provide DevSecOps and security enablement sessions , architectural guidance, and technical consulting for internal customers.
6. Operational Excellence
§ Ensure reliability, resilience, and secure operations of AI agent platforms.
§ Optimize cost, performance, monitoring, and security across environments.
§ Drive automation, continuous improvement, and operational maturity through DevSecOps practices.
Core AI / Agentic Skills
§ Experience with AI agents, LLM -based architectures, and multi -agent systems.
§ Strong understanding of:
§ RAG (Retrieval -Augmented Generation), Vector DBs o Prompt orchestration, tool calling, and function execution
§ Memory management and long -context handling
§ Multi-agent collaboration patterns and interoperability standards
AI / ML Engineering
Experience with:
§ Python, FastAPI , TypeScript/Node (preferred but not mandatory)
§ Model serving (Databricks, SageMaker endpoints, Azure ML endpoints)
§ MLOps pipelines (CI/CD, deployment automation, model lifecycle management)
§ Evaluation frameworks for agents and LLMs
DevSecOps & Platform Engineering (Primary Skill)
Strong expertise in:
§ CI/CD pipeline design with integrated security scanning and policy enforcement
§ Infrastructure as Code (Terraform, Bicep, CloudFormation)
§ Secure build, test, release, and deployment automation
§ Continuous compliance, auditability, and traceability
§ Observability (logging, metrics, tracing) for AI systems
Cloud & Infrastructure
• Expert in AWS and Azure .
• Hands-on with: o Kubernetes (AKS, EKS)
§ Serverless (Lambda, Step Functions)
§ API Gateway and microservices
§ VPC networking, peering, firewalls
§ Identity & Access Management (IAM/RBAC)
Data Architecture
Familiar with:
§ Data lakes, RDBMS, DynamoDB
§ ETL/ELT workflows
§ Data governance, cataloging, and metadata handling
§ Secure access to enterprise data sources
Security
Deep understanding of:
§ OAuth2, JWT, secrets management
§ SOC2, GDPR, privacy controls
§ Threat modeling for agentic and AI systems
§ Safe prompting, guardrails, filtering, and toxicity check
