Enterprise Architect — AI & Agentic Platform
Stelfox is partnering with a U.S Fintech who is on a pathway to buildout Agentic AI capabilities across their SaaS financial solutions portfolio. Their vision is an Agentic platform that surfaces intelligent insights, executes autonomous workflows, and enables natural language interaction with complex financial data, all embedded natively into risk & performance management tools used key client stakeholders/users daily.
This platform will serve as the core AI backbone for the company's product suite being deployed with enterprise customers.
As Enterprise Architect, you will report directly to the CTO to:
Act as the senior technical authority defining how AI and agentic capabilities are designed, built and embedded across the company's product suite.
Your role will involve both defining the patterns and proving them out.
You will spend 50% of your time on architecture, standards and cross‑team alignment, with the other 50% hands‑on building prototypes / PoCs and reference implementations that the Platform services pod productionises and domain.
You will bridge between "what is possible with AI" and "what is shipped in product".
What this role is NOT:
Managing people (you will influence via technical authority and working code).
Operate as a strategy‑only advisor who focused on slide decks.
Build production systems end‑to‑end (platform engineering takes care of this).
What this role will be:
Defining Agentic Platform architecture and the patterns product teams follow.
Build working prototypes that prove architectural decisions before teams invest in production builds.
Establish the API and MCP server strategy that exposes products to agentic consumption.
Provide architectural oversight of ML/statistical modeling for analytics use cases.
Be the technical tiebreaker on AI architectural decisions across the entire organisation.
Required Qualifications
Experience
10+ years in software architecture/engineering, with 4+ years focused on AI/ML systems in production.
Critical: Personally designed and built agentic or AI‑orchestrated systems beyond basic RAG/chatbot implementations — multi‑step reasoning, tool use, autonomous workflows.
Critical: Defined API strategies for complex, multi‑product platforms — not just consumed APIs, but designed the contracts, versioning, and developer experience.
Production experience with LLM‑based systems: prompt engineering, orchestration frameworks, evaluation/observability, guardrails, cost optimization.
Demonstrated ability to translate architectural vision into working prototypes that teams can adopt.
Track record of influencing engineering direction through technical authority rather than positional authority.
Technical Expertise
Agentic/AI Platforms: Deep hands‑on experience with orchestration frameworks (LangGraph, LangChain, CrewAI, Semantic Kernel, or equivalent); agent design patterns; tool‑use architectures; MCP or similar protocols.
LLM Infrastructure: Azure AI Foundry / Azure OpenAI, AWS Bedrock, or equivalent; model selection and routing; token economics and cost management at scale.
Observability & Evaluation: LangSmith, Braintrust, custom evaluation frameworks; agent tracing; quality metrics for non‑deterministic systems.
API Architecture: RESTful and event‑driven API design; versioning strategies; contract testing; developer experience.
ML/Data Science Foundations: Sufficient depth to guide ML architecture decisions — regression modeling, time series analysis, model validation, MLOps patterns (you don’t need to be a data scientist, but you need to know enough to architect the systems they depend on).
Patterns for AI workloads.
Languages
Proficiency in Python + one of TypeScript/C# (our stack is .NET, TypeScript/Angular, Python).
#J-18808-Ljbffr