E2E AI SME Engineer

Univedge Consulting LLC is hiring an E2E AI SME Engineer-cum-Product Owner to lead the strategy, development, deployment, and continuous improvement of enterprise AI products. The role combines hands-on AI engineering with product ownership, stakeholder leadership, production operations, and responsible AI governance.

Responsibilities:

• Own the end-to-end AI product lifecycle, including discovery, vision, business case, roadmap, backlog, delivery, launch, adoption, and continuous improvement
• Partner with business, data, engineering, security, legal, and operations stakeholders to define user journeys, functional and non-functional requirements, acceptance criteria, and measurable outcomes
• Prioritize features and technical enablers based on customer value, feasibility, risk, cost, and strategic alignment; lead sprint planning, reviews, and release readiness
• Design and build production-grade AI applications using Python, APIs, LLM frameworks, vector databases, and cloud-native services
• Develop LLM, RAG, agentic AI, fine-tuning, prompt engineering, and tool-calling solutions for enterprise use cases
• Create reference architectures, reusable components, technical prototypes, and coding standards while contributing directly to critical implementation work
• Define data ingestion, chunking, embedding, retrieval, orchestration, and model-selection approaches across structured and unstructured data
• Establish offline and online evaluation frameworks covering quality, relevance, groundedness, hallucination, latency, cost, safety, and business KPIs
• Lead experimentation, model comparison, prompt and retrieval optimization, red-teaming, and human-in-the-loop validation
• Lead secure deployment of AI services across development, test, and production environments using CI/CD, containers, infrastructure as code, and automated testing
• Implement observability for model and application performance, data drift, prompt and response traces, token usage, latency, reliability, and cost
• Own production readiness, incident response, root-cause analysis, release governance, vendor coordination, and service-level objectives
• Embed privacy, security, responsible AI, explainability, access control, auditability, and regulatory requirements throughout the product lifecycle
• Identify and mitigate risks related to sensitive data, prompt injection, data leakage, model misuse, bias, harmful output, and third-party models
• Communicate product decisions, trade-offs, architecture, progress, risks, and value realization to technical and executive audiences
• Drive pilots, user feedback, change management, training, adoption metrics, and scale-up plans; mentor engineers and promote AI engineering best practices