AI Platform Engineer

Overview:

Mandatory skills

• The platform should enable teams to build, deploy, and operate AI agents using a variety of development approaches and frameworks while providing a standardized enterprise foundation for orchestration, integration, governance, security, and observability.

• The engineer should have experience designing agentic workflows and multi-agent orchestration, including both event-driven and workflow-based patterns. The platform should support scalable communication and coordination between agents, enterprise systems, tools, APIs, and diverse data sources.

• A key responsibility will be establishing enterprise-grade observability across the platform, including centralized instrumentation, tracing, operational metrics, performance monitoring, error tracking, and visibility into agent execution and behaviour.

• The platform should also provide a centralized gateway for AI model interactions, agent tools, and external services, ensuring consistent security, governance, authentication and authorization, access control, throttling, monitoring, and policy enforcement.

• The ideal engineer will have experience building scalable, extensible, and technology-agnostic AI platforms that allow different teams and business domains to develop agents independently while adhering to common enterprise standards for interoperability, security, governance, reliability, and operational management.

• The role should also focus on building production-grade AI engineering capabilities, including Agent Harness Engineering, automated and closed-loop evaluation, feedback loops, prompt and model evaluation, observability, guardrails, resiliency, and continuous improvement mechanisms. The goal is to create a reusable platform where multiple business domains can build, deploy, monitor, and operate autonomous agents using standardized enterprise patterns rather than creating isolated agent solutions.

Skills:

RAG Pipeline, LLM, Enterprise Applications, AgenticAI, Python, Azure