AI/ML Architect

Summary

The ideal candidate is a hands-on AI Architect with deep expertise in LLM-based systems, agentic architectures, and AWS-native platforms, combined with strong experience in guardrails, evaluation frameworks, and cost optimization. This role requires a balance of advanced technical depth and architectural leadership to deliver scalable, production-grade AI platforms

• Enterprise AI & LLM Architecture (8+ years total, 4+ years in AI/ML architecture)
• Architect and deliver enterprise-scale AI systems leveraging Large Language Models (LLMs), including Retrieval-Augmented Generation (RAG), fine-tuning strategies, and advanced agentic architectures.
• Design multi-agent systems capable of tool orchestration, planning, reasoning, and context management for complex, distributed workflows.

• LLM Guardrails & Responsible AI (1+ years)
• Design and implement guardrails for LLM applications, including content filtering, prompt injection mitigation, hallucination reduction, and policy enforcement.
• Experience with safety frameworks, human-in-the-loop validation, and governance controls to ensure compliance, reliability, and ethical AI usage.

• LLM Cost Optimization & Performance Engineering (1+ years)
• Optimize LLM workloads for cost, latency, and throughput, including model selection strategies, prompt engineering efficiency, caching mechanisms, batching, and token optimization.
• Experience implementing cost monitoring, usage controls, and scalable inference architectures to balance performance with operational efficiency.

• Evaluation Frameworks & Model Quality (Evals) (1+ years)
• Develop and operationalize evaluation (evals) frameworks for LLMs, including automated benchmarking, prompt evaluation, response scoring, and regression testing.
• Experience with both offline and online evaluation techniques, including human evaluation pipelines, A/B testing, and continuous feedback loops to improve model quality and reliability.

• AWS Bedrock & Agentic Frameworks (2+ years hands-on)
• Hands-on experience with Amazon Bedrock, including foundation model evaluation, prompt engineering, orchestration, and model lifecycle management.
• Experience with AgentCore or similar agent frameworks for building composable, stateful, and tool-augmented AI agents with memory and execution control.

• Advanced Python & API Engineering (6+ years)
• Expert-level proficiency in Python with deep experience across AI/ML ecosystems (PyTorch, TensorFlow, Hugging Face, LangChain, or equivalent).
• Strong experience building scalable, production-grade APIs using FastAPI, including async processing, model inference optimization, and microservices architecture.

• CI/CD and MLOps for AI Systems (5+ years)
• Design and implement robust CI/CD pipelines for AI applications, including automated model testing, validation, reproducibility, and staged rollouts.
• Experience with MLOps frameworks, model versioning, feature stores, and continuous monitoring using tools such as GitHub Actions, Jenkins, AWS CodePipeline, and Terraform/CloudFormation.

• AWS Cloud Architecture (2 + years)
• Basic expertise in Amazon VPC design, including subnetting strategies, private networking, security segmentation, and high-availability architecture for AI workloads.
• Hands-on experience with AWS Lambda and event-driven patterns for scalable, serverless inference pipelines and integration with AWS services (API Gateway, SQS, Step Functions).

• Distributed Team Leadership & Global Delivery (5+ years)
• Proven experience leading distributed engineering and AI development teams across multiple time zones, ensuring high productivity and alignment.
• Strong track record of implementing agile delivery models, managing cross-regional collaboration, and driving continuous delivery across globally dispersed teams.