Senior AI Transformation Engineer

شرح موقعیت

We’re looking for a Senior AI Transformation Engineer who is responsible for identifying, prototyping, and delivering production-grade AI solutions and automations that transform internal business workflows, eliminating repetitive friction, and driving measurable operational efficiency. This role requires strong production software engineering fundamentals, deep expertise in LLM application architecture, tool calling, agentic orchestration, and RAG systems, alongside the business maturity to map processes, evaluate feasibility versus cost, and choose pragmatic software or deterministic automation whenever appropriate. Key Responsibilities Discovery & Opportunity Prioritization: Partner with department leads and process owners to map internal workflows, identify repetitive manual tasks, and prioritize initiatives based on business impact, feasibility, cost, risk, and ROI. Requirements & Solution Architecture: Translate ambiguous business challenges into precise technical designs, delivery roadmaps, boundary definitions, and quantifiable success metrics. AI Application & Agent Development: Build robust AI-powered internal tools, automations, and multi-step agents using LLM APIs, function/tool calling, structured outputs, prompt engineering, and context engineering. RAG & Enterprise Integration: Implement knowledge retrieval and RAG pipelines (chunking, vector search, reranking) while seamlessly integrating AI services with internal APIs, databases, ERPs, and legacy business platforms. Prototyping & Production Hardening: Validate technical assumptions through rapid, lightweight proof-of-concepts, transitioning validated experiments into secure, scalable, and maintainable production software. Evaluations, Guardrails & Governance: Build evaluation frameworks, regression test suites, guardrails, role-based access controls (RBAC), and human-in-the-loop validation mechanisms to mitigate hallucinations, privacy risks, and multi-step failures. Monitoring & Observability: Track operational metrics including response latency, output quality, token usage, and infrastructure costs, actively troubleshooting edge cases, schema violations, and upstream model drift. Adoption, Documentation & Enablement: Measure user adoption and time-saved metrics, document integration patterns and playbooks, and conduct onboarding sessions to help internal teams comfortably adopt AI-enhanced workflows. Qualifications & Requirements 5+ years of hands-on software engineering experience building and maintaining production-grade services and applications. Proven track record of developing and deploying production AI solutions using LLM APIs, agentic architectures, tool calling, structured outputs, and RAG pipelines. Strong backend and integration fundamentals, including robust API design, database schemas, authentication flows, error handling, and idempotency. Strong grasp of evaluation methodologies, tracing, prompt/context engineering, and output validation safeguards. High degree of ownership, stakeholder management, and communication skills to explain technical tradeoffs and operational changes to non-technical business leaders. Pragmatic engineering mindset: ability to recognize when deterministic software or simple rules-based automation is superior to an LLM. Preferred Qualifications (Nice to Have) Hands-on experience with agent orchestration frameworks (OpenAI Agents SDK, LangGraph, Mastra, or similar) and Model Context Protocol (MCP). Experience with AI observability, tracing, and evaluation platforms (e.g., Langfuse, Arize Phoenix, Braintrust, or TruLens). Experience with enterprise document processing, chunking strategies, semantic search, and hybrid vector/lexical retrieval. Exposure to event-driven architectures, queueing systems, and modern CI/CD automation pipelines. What Makes This Role Unique This role is not about training speculative models or building toy demos; it sits directly at the intersection of production software engineering and company-wide digital transformation. You will hold end-to-end technical autonomy to identify high-friction bottlenecks across business units, deploy dependable agentic workflows into production, and directly accelerate the speed, quality, and intelligence of how the entire company operates.