Gen AI Architect

Position: Gen AI Architect

Sourcebae is seeking a highly experienced Gen AI Architect to design and implement advanced Generative AI and multi-agent systems. This role focuses on architecting agentic workflows, RAG architectures, and robust MLOps/GenAIOps practices using cutting-edge Google Cloud Platform technologies.

Key Responsibilities

• Generative AI & Multi-Agent Systems: Design and build multi-agent systems leveraging the Google Agent Development Kit (ADK) and orchestration frameworks like LangGraph or CrewAI. Implement complex interaction patterns (Sequential, Parallel, Iterative Refinement) where specialized agents (e.g., Researcher, Coder, Evaluator) collaborate to solve non-linear tasks. Design high-performance Retrieval-Augmented Generation (RAG) architectures using Vertex AI Search and BigQuery Vector Search.
• ML Training & Production Deployment: Train and deploy machine learning models such as Random Forests, XG Boost, LightGBM, and various forecasting models on Vertex AI. Deploy models for low-latency online inference and high-throughput batch processing using Vertex AI Endpoints and Cloud Run.
• MLOps & GenAIOps: Build end-to-end Vertex AI Pipelines (Kubeflow-based) for automated data ingestion, training, evaluation, and deployment. Implement Model Monitoring to track data drift, prediction drift, and GenAI-specific metrics like hallucination rates and grounding scores. Establish Continuous Integration, Continuous Deployment, and Continuous Training (CI/CD/CT) workflows to ensure model reliability and seamless updates.

Required Qualifications

• Education: Bachelor’s or Master’s degree in Computer Science, Data Science, or a related field.
• Experience: 7+ years in software/data engineering, with 3+ years specifically focused on architecting and deploying AI/ML solutions in production.
• GCP Mastery: Deep hands-on experience with Vertex AI (Model Garden, Pipelines, Agent Builder, Feature Store) and BigQuery.
• Agentic Frameworks: Proven experience building multi-agent systems and managing state/memory across complex agent loops.
• Programming: Expert-level Python skills and experience with the modern AI stack (Scikit-Learn, PyTorch, TensorFlow, LangChain, Hugging Face).
• DevOps/MLOps: Proficiency with Docker, Kubernetes.

Key skills/competency

• Generative AI
• Multi-Agent Systems
• Retrieval-Augmented Generation (RAG)
• Vertex AI
• MLOps
• GenAIOps
• Google Cloud Platform (GCP)
• Python
• Kubernetes
• TensorFlow/PyTorch