AI Consultant
Location: Remote from Spain (an indefinite Spanish employment contract)
Our client is the fastest-growing global manufacturing company. An international corporation with over a hundred years of history, internationally recognized brands and Reduced-Risk Products.
Intellias' mission is to support its strategy and efforts in the Digital and e-commerce space (e-commerce and other apps mobile apps, payment gateways, loyalty system, search engine, employee management, identity management, etc.).
A newly conceptualized Digital Eco System is comprised of a set of capabilities including an online shop & website, linking online & offline, customization & personalization, engagement & membership, digital product & services main differences.
Requirements:
* 5+ years in MLOps/platform architecture or adjacent roles, with shipped AI systems
* Proficient Python and strong software engineering principles
* Deep experience with at least one major cloud (AWS/Azure/GCP) and platform engineering (containers, Kubernetes, IaC such as Terraform)
* Experience in designing and guiding scalable machine learning pipelines for model training, validation, and deployment
* Proven CI/CD design for GenAI/ML (evaluation gates, versioning, canary, rollback) and collaboration with security/governance stakeholders
* Sound judgement selecting RAG/vector and provider stacks based on performance, cost, compliance, and portability
* Agent orchestration frameworks (e.g., LangGraph/Semantic Kernel) and tooling protocols (e.g., MCP)
* Experience operationalizing multi-agent systems (tools/routing/memory/guardrails, human-in-the-loop)
* Process automation and enterprise integrations
* Excellent communication and interpersonal skills to collaborate effectively with cross-functional teams, stakeholders' leadership
* Upper-intermediate level of English
Nice to have:
* Master or higher degree in Computer Science, Engineering, or related field
* On-prem LLM deployments; performance and cost tuning with caching and model routing
* AI safety, policy, and compliance experience in sensitive environments
* Public speaking and enablement and building reusable accelerators
* Domain exposure in automotive, retail, manufacturing, healthcare, energy, finance, or telecom
Responsibilities:
1. Architecture Review & Production Readiness Assessment
• Evaluate AI/ML and LLM solution architectures to ensure they are scalable, secure, and aligned with enterprise patterns.
• Assess MLOps/LLMOps pipelines, model serving infrastructure, data flows, and integration points.
• Identify architectural risks or gaps and propose mitigation strategies.
2. Compliance & Standards Validation
• Verify that all AI development activities follow JTI’s internal development standards, documentation rules, and operational guidelines.
• Ensure compliance with model governance, lifecycle management, versioning, and traceability requirements.
• Check adherence to security, privacy, and data handling policies.
3. Technical Quality Assurance
• Perform in‑depth technical code reviews, configuration reviews, and environment checks.
• Validate model performance metrics, evaluation methodology, drift controls, and monitoring strategies.
• Review model explainability, responsible‑AI controls, and risk assessment outputs.
4. Pre‑Deployment Validation
• Conduct formal readiness reviews before solutions are promoted to production.
• Provide clear recommendations for required fixes, improvements, or optimization.
• Approve or block deployment based on technical quality and compliance.
5. Documentation & Reporting
• Produce detailed review reports summarizing findings, gaps, and actionable guidance.
• Maintain traceability of assessments across multiple projects throughout 2026.
6. Cross‑Team Collaboration
• Collaborate with engineering, data science, architecture, and product teams to clarify requirements and ensure alignment.
• Participate in technical workshops and solution walkthroughs.
7. Advisory & Best Practices Enablement
• Advise teams on AI/ML and LLM best practices, including architecture, operations, MLOps, evaluation, and productionization.
• Help standardize review processes and improve internal frameworks when needed.