<p>Job Duty Descriptions</p> <ul> <li>Lead architecture for GenAI/LLM solutions using patterns such as RAG, tool/function calling, and agentic workflows. </li><li>Design secure enterprise knowledge integration and define data access/segregation controls. </li><li>Establish Responsible AI, privacy, and security controls for GenAI. </li><li>Provide technical leadership across stakeholders; document architecture decisions, risks, and roadmaps; guide teams on performance and cost optimization. </li><li>Define AI/ML target architecture and reference patterns aligned to business strategy, enterprise standards, and regulatory requirements. </li><li>Translate business use cases into end-to-end solution designs across data, ML, application, integration, and infrastructure layers. </li><li>Establish technical guardrails (security, privacy, scalability, performance, observability) for AI/ML platforms and workloads. </li><li>Review and approve architecture/designs; document decisions, trade-offs, and standards; mentor teams on best practices. </li><li>Design and govern the AI/ML platform and MLOps lifecycle (data ingestion, training, validation, deployment, monitoring, retraining). </li><li>Define CI/CD standards for ML (model versioning, reproducible pipelines, automated testing, approvals, rollback). </li></ul> <p>Day-to-day Tasks</p> <ul> <li>Prototype RAG/agent workflows; tune chunking/retrieval. </li><li>Implement guardrails: input/output filtering, prompt templates, tool whitelisting, rate limits, and secrets management; perform threat modeling and red teaming. </li><li>• Run architecture workshops with engineering teams to capture requirements and constraints. </li><li>Create solution blueprints, component diagrams, and integration flows; define non-functional requirements (NFRs). </li><li>Conduct design reviews, security reviews, and performance reviews; provide actionable remediation guidance. </li><li>Maintain architecture repository and reusable templates. </li><li>Define pipeline templates and guardrails; onboard teams to standard training/deployment workflows. </li><li>Set up model registries, feature stores, and experiment tracking standards; enforce naming/versioning conventions. </li><li>Implement model monitoring. </li><li>Create runbooks and release checklists; support go-lives and post-release validation. </li><li>Perform periodic platform health and cost reviews; recommend tuning and right-sizing actions. </li></ul> <p>Technology and/or Software used for Duty</p> <ul> <li>Azure OpenAI / Vertex AI / Bedrock </li><li>LangChain / LlamaIndex </li><li>Vector DB </li><li>C ontent safety/guardrails (Azure AI Content Safety, policy filters) </li><li>Python, APIs, Git/Azure DevOps, Docker/Kubernetes </li><li>Azure/AWS/GCP </li><li>Kubernetes </li><li>MLflow </li><li>Kubeflow / Airflow </li><li>Git / Azure DevOps </li><li>Docker / Kubernetes </li><li>Prometheus / Grafana </li></ul> <p>Education</p> <p>Bachelor's degree</p> <p>Salary Range: $180000 - $230000 a year</p> <p>#LI-JH1</p>