What are the responsibilities and job description for the AI Platform Architect position at SymphonyAI?
Job Description
We are seeking an experienced AI Architect to design, govern, and scale end-to-end AI solutions that deliver measurable business outcomes. This role sits at the intersection of data, machine learning, engineering, and product, translating business needs into robust, secure, and scalable AI architectures.
The AI Architect will define reference architectures, select platforms and tools, and guide teams in building production-grade AI systems across the enterprise.
Key Responsibilities
Platform Architecture & Vision
Cloud & Platform Engineering
We are seeking an experienced AI Architect to design, govern, and scale end-to-end AI solutions that deliver measurable business outcomes. This role sits at the intersection of data, machine learning, engineering, and product, translating business needs into robust, secure, and scalable AI architectures.
The AI Architect will define reference architectures, select platforms and tools, and guide teams in building production-grade AI systems across the enterprise.
Key Responsibilities
Platform Architecture & Vision
- Own the end-to-end architecture for the AI platform, spanning:
- Agent frameworks and orchestration layers
- Semantic and knowledge graph foundations
- Data and signal ingestion fabric
- Model, reasoning, and tool-execution services
- Product and solution enablement layers
- Establish modular, extensible reference architectures enabling rapid product and solution development.
- Drive architectural consistency across teams building on AI Platform.
- Agentic & Knowledge-Driven AI Systems
- Architect agent-based systems capable of reasoning, planning, retrieval, and execution across enterprise workflows.
- Design hybrid AI architectures combining:
- LLMs and multi-model stacks
- Knowledge graphs and ontologies
- Vector retrieval and semantic search
- Deterministic services and enterprise APIs
- Lead the evolution of CINDE’s semantic layer and retail knowledge foundation.
- Solution Architecture & Business Enablement
- Partner with Product, Engineering, and Business leaders to translate strategy into scalable technical systems.
- Architect AI solutions across retail and CPG domains, including:
- Forecasting, demand intelligence, and optimization
- Price, promotion, and assortment intelligence
- Shopper personalization and retail media
- Store, shelf, and inventory intelligence
- Enterprise revenue and decision automation
- Ensure architectures directly support revenue growth, product velocity, operational efficiency, and customer impact.
- AI Platform Engineering, MLOps & LLMOps
- Define CINDE standards for:
- Model lifecycle management
- Agent deployment and orchestration
- Prompt, workflow, and tool governance
- Experimentation and evaluation pipelines
- Design scalable MLOps / LLMOps / AgentOps foundations:
- CI/CD for AI and agent workflows
- Observability, telemetry, and quality measurement
- Versioning, monitoring, drift detection, and retraining
- Governance, Security & Responsible AI
- Embed enterprise-grade security, privacy, and compliance into CINDE architecture.
- Define and enforce Responsible AI frameworks across the platform:
- Explainability, traceability, and auditability
- Bias, safety, and risk controls
- Regulatory and customer-facing compliance readiness
- Partner closely with Security, Legal, and Compliance leaders.
- Technical Leadership & Influence
- Serve as a technical north star across product and engineering organizations.
- Mentor senior engineers, architects, and data scientists.
- Influence platform decisions across multiple business units without direct authority.
- Continuously assess emerging technologies and translate them into advantage.
Cloud & Platform Engineering
- Deep experience with AWS, Azure, or GCP AI platforms
- Kubernetes, containerized AI workloads, and distributed systems
- Infrastructure as Code and environment automation
- Enterprise data lakes and lakehouse platforms
- Streaming and real-time signal architectures
- Strong distributed data processing background
- Knowledge graph platforms, semantic modeling, and ontologies
- Expert-level Python
- Production ML frameworks (PyTorch, TensorFlow, scikit-learn)
- Agent frameworks and orchestration platforms
- Multi-model system design
- Commercial and open-source LLM ecosystems
- RAG and hybrid retrieval architectures
- Vector databases and embedding systems
- Fine-tuning, evaluation, and prompt lifecycle managemen
- MLflow, Kubeflow, or equivalent platforms
- CI/CD for AI workloads
- Model and agent observability, testing, and governance