Senior Amazon Bedrock Engineer — AI Infrastructure & Cost Optimization

Location: Remote (U.S.) | Type: Full-time or Contract

Company: Aesthetics360 (A360)

About A360

Aesthetics360 is an AI-native platform transforming elective medicine.
We combine domain-trained models, live transcription, and automated marketing into one continuous intelligence system.

Every workflow — from consult to care plan — runs on AWS, with Amazon Bedrock at the core of our generative-AI architecture.
We’re scaling fast and need a Bedrock specialist who can make our inference layer faster, cheaper, and smarter.

The Role

You’ll be the owner of Bedrock orchestration inside A360 — optimizing prompt routing, model performance, and API utilization while pairing that with DevOps rigor and cost governance.
You’ll design scalable, secure, and financially efficient ways to serve multi-model workloads (Claude, Titan, Mistral, Llama 3 via Bedrock) to thousands of real-time users.

What You’ll Do

• Architect, deploy, and optimize Bedrock-based LLM pipelines powering transcription, summarization, and consultation workflows.
• Implement multi-model orchestration and intelligent routing for cost/performance balance (e.g., Haiku → Claude 3 → Titan RAG chains).
• Build automated monitoring, logging, and cost dashboards for token usage, latency, and throughput (CloudWatch + Cost Explorer + custom metrics).
• Engineer caching and batching strategies to reduce Bedrock inference cost per request.
• Manage CI/CD for prompt chain deployment using AWS CDK, Fargate, and Lambda.
• Collaborate with AI R&D to profile prompt performance, fine-tune templates, and benchmark models.
• Enforce FinOps best practices — budget alerts, anomaly detection, and savings-plan recommendations.

• Ensure enterprise-grade uptime, compliance, and security across Bedrock environments.

What You Bring

• Deep expertise in Amazon Bedrock (multi-model orchestration, API integration, cost tuning).
• 5 + years hands-on AWS architecture or DevOps experience (ECS, Lambda, Aurora Serverless, API Gateway).
• Proficiency with AWS CDK / Terraform, CloudWatch, and Cost Explorer APIs.
• Experience optimizing token budgets, latency, and context-window trade-offs in LLMs.
• Familiarity with LangChain, SageMaker, or Supabase is a plus.
• Strong Python and shell scripting skills; ability to build internal CLI tools for cost and performance audits.
• Mindset: equal parts AI architect and FinOps engineer — you design with performance and price in mind.