AI Architect (6 months contract)

Overview

Full-Time

Role Overview

Key Responsibilities

• Define AI strategy and reference architectures for generative AI, agentic systems, RAG, and predictive ML.
• Set model strategy: build-vs-buy, open-weight vs. hosted, multi-model routing, and a framework for evaluating new models as they ship.
• Architect the data and retrieval foundation — lakehouse, feature stores, vector stores, and lineage — that AI systems depend on.
• Design the LLMOps/MLOps platform: experiment tracking, model and prompt registries, automated eval gates, and observability.
• Lead AI security: threat modeling for prompt injection and data exfiltration, and OWASP LLM Top 10 mitigations.
• Own responsible AI and governance — risk classification, human oversight, auditability, and alignment with the EU AI Act, NIST AI RMF, and ISO/IEC 42001.
• Establish FinOps for AI: cost models, token budgeting, and unit economics per use case.
• Provide technical leadership, review designs, mentor engineers, and align stakeholders from engineering to the executive team.

Candidate Requirements

• Demonstrated experience delivering AI solutions in an enterprise or client-facing/consulting environment.
• Prior experience at management consulting firms and/or Big Tech is an advantage.
• Bachelor's or Master’s degree in Computer Science, Engineering, Information Technology, or a related field.
• 8+ years in software/data engineering, with 3+ years architecting AI/ML or large-scale data systems.
• A track record of designing and delivering production AI systems end-to-end, ideally including GenAI/LLM workloads.
• Deep understanding of LLM and ML system design — RAG, agents, fine-tuning trade-offs, evaluation, serving, and the full MLOps/LLMOps lifecycle.
• Strong cloud architecture skills (Azure, AWS, or GCP) plus a modern data platform (Databricks or Snowflake).
• Expertise in AI security, responsible AI, and data governance, with awareness of relevant regulations.
• Excellent communication — able to influence both engineers and executives.
• Hands-on enough to prototype, review code, and earn engineering teams’ trust.
• Nice to have:
• Architecting AI for regulated or sovereign environments (finance, healthcare, government, GCC/UAE).
• Enterprise architecture frameworks (TOGAF) and well-architected reviews.
• Defining golden-path tooling for internal AI development.
• Running structured vendor PoCs and bake-offs.

Tools & Technologies You’ll Work With

• Patterns: RAG, agentic/multi-agent, event-driven, lakehouse/medallion; reusable blueprints.
• Model strategy: multi-model routing and gateways (LiteLLM, Portkey); Claude, GPT, Gemini, Llama, Mistral.
• Platforms: Azure AI Foundry, AWS Bedrock, Vertex AI, Databricks Mosaic AI.
• Data foundation: Delta Lake/Databricks, Snowflake, Unity Catalog, vector stores, knowledge graphs.
• LLMOps/MLOps: MLflow, LangSmith/Langfuse, model and prompt registries, eval-in-CI.
• Security & governance: OWASP LLM Top 10, NIST AI RMF, ISO/IEC 42001, EU AI Act, RBAC/ABAC, secrets management.
• Infra & observability: Kubernetes, Terraform/OpenTofu, OpenTelemetry; token/cost dashboards.