AI Architect

About the role

Job Overview The AI Architect will design and implement end-to-end architectures for Generative AI and AI/ML solutions, working under the guidance of the Director Enterprise Systems to bring the organization's AI strategy to life. Our EdTech products and services touch millions of students and educators every year so our AI solutions must be guided by our purpose to help every teacher and student feel seen, valued and supported. This role is deeply technical, hands-on, and delivery-focused. The ideal candidate has strong experience with data-driven AI, GenAI architectures (especially RAG), cloud-native patterns, and MLOps/LMMOps tooling. They will be a key contributor to the build‑out of enterprise AI platforms and accelerators. The AI Architect translates architectural standards into real solutions, builds reusable components, partners closely with data engineering and product teams, and ensures AI projects are secure, scalable, compliant, and production‑ready. Job Responsibilities: 1. Solution Architecture & Delivery Design secure end-to-end AI solution architectures including data ingestion, model training, inference pipelines, orchestration flows, and integration with downstream systems. Implement architectures defined by the Director Enterprise Systems, ensuring alignment with standards, patterns, and platform strategy. Build GenAI solutions using RAG, vector search, grounding strategies, prompt orchestration, and model evaluation frameworks. Create and maintain high-quality HLD/LLD documentation, sequence diagrams, and data flows for AI workloads. Perform hands-on technical proofs of concept, evaluate models/tools, and convert prototypes into production-grade systems. 2. AI Engineering Enablement Build secure reusable code templates, libraries, and patterns for deployment, evaluation, and monitoring of workloads. Partner with engineers to integrate AI components into pipelines, data products, and operational workflows. Implement model lifecycle management: versioning, experimentation tracking, registry integration, automated deployment. 3. Data & Retrieval Architecture Architect federated data access patterns for AI, integrating multiple source systems (data lakes, warehouses, content repositories, SaaS platforms) into cohesive retrieval pipelines. Design data pipelines that support AI use cases: feature engineering, embedding generation, chunking strategies, and retrieval flows. Implement and optimize vector DB schemas, embeddings, and hybrid search (keyword + semantic) patterns. Ensure data quality, lineage, access controls, and privacy protections align with enterprise requirements. 4. Governance, Security & Compliance Partner with Security and Privacy teams to ensure AI solutions align with applicable regulations and standards, translating policy requirements into enforceable technical controls. Apply responsible AI principles, ensuring solutions include safety, bias mitigation, grounding, and hallucination safeguards. Establish guardrails for model training and prompt usage, including restrictions on sensitive data ingestion and prevention of model leakage. Implement enterprise-approved security patterns (private endpoints, tokenization/MPC, encryption, IAM roles, network controls). Conduct architecture reviews, risk assessments, and model evaluations as part of deployment readiness. 5. Collaboration & Communication Collaborate with product managers, engineers, and business stakeholders to refine requirements and translate them into technical specifications. Provide clear technical guidance, mentoring, and code reviews for teams using AI services. Communicate trade-offs, limitations, and risks of different AI approaches to both technical and non-technical audiences. Maintain a strong business perspective to ensure systems are implemented in ways that support operational goals and user needs. Ensure solutions meet high standards of quality, with successful delivery driven by thorough testing and validation practices. Provide technical guidance to other engineering team members, fostering growth and knowledge sharing. Job Requirements: 5+ years of experience in engineering and architecture roles, operating at enterprise scale and demonstrating sustained ownership of complex data and AI enabled platforms. Hands-on experience designing and implementing: LLM-based solutions (RAG, tool use, prompt workflows, fine-tuning where appropriate). Traditional ML models and pipelines. Cloud AI services (Azure OpenAI/AI Studio, AWS Bedrock/SageMaker, GCP Vertex AI). Vector databases & search (Azure AI Search, Pinecone, Weaviate, OpenSearch, pgvector). Data pipelines supporting AI (Spark, Databricks, Synapse, Snowflake, dbt, Airflow). Strong understanding of MLOps/LMMOps practices including model registry, CI/CD for ML, monitoring, and evaluation. Proficiency in Python and familiarity with key AI frameworks (LangChain/LangGraph, Semantic Kernel, PyTorch or TensorFlow, MLflow). Working knowledge of security