Principal Technical Program Manager, AWS Mantle
Summary
Principal TPM defining and driving global AI inference programs for AWS Mantle, Amazon Bedrock’s next-gen distributed inference engine, shaping infrastructure scale and security at VP level.
Are you a strategic technical program leader who thrives in high-ambiguity environments where the problem itself may not yet be defined? We are seeking a Principal Technical Program Manager to drive the delivery of significantly complex, cross-organizational programs for AWS Mantle—the next-generation distributed inference engine powering Amazon Bedrock. In this role, you will operate with complete independence across VP organizations, defining program strategy, driving clarity where none exists, and making trade-offs between business opportunity, resources, and long-term sustainability. Your work will be strategic in nature, shaping how we scale AI inference infrastructure to serve millions of customers globally.
- Define and own very large programs spanning multiple engineering teams, product organizations, and VP orgs—driving cross-functional alignment on Mantle's most critical technical initiatives
- Operate in deep ambiguity where the business and architectural strategy may not yet be defined; drive clarity and create the path forward
- Identify risks and opportunities in technical strategies, architectures, and engineering organization structures, influencing decision-makers at the VP level
- Make strategic trade-offs between business opportunity, resource allocation, and system sustainability while remediating critical and endemic problems
- Effectively force-multiply across the Mantle organization—accelerating delivery, managing escalations, and aligning teams toward simple, coherent approaches
Key job responsibilities
As a Principal TPM on the Mantle team, you will be the strategic connective tissue across our distributed inference platform—orchestrating the delivery of programs that span model onboarding, capacity management, security architecture, and global API availability. You will not simply manage timelines; you will define missions, set objectives, analyze data, and influence resource allocation at the highest levels.
- Own end-to-end program strategy for Mantle's most complex cross-organizational initiatives, including global model onboarding pipelines, Zero Operator Access security programs, and multi-region capacity scaling
- Drive technical architecture discussions with Principal Engineers and SDMs, identifying dependencies, scaling factors, boundary conditions, and risks across distributed systems serving ML inference at global scale
- Represent and advocate for customers during executive-level prioritization and planning, owning roadmap strategy and ensuring programs deliver measurable outcomes
- Create and optimize cross-org structures and mechanisms that align teams toward coherent delivery approaches—eliminating unnecessary process and streamlining execution
- Actively develop the TPM community through mentorship, promotion assessments, and raising the bar on program management excellence across Amazon Bedrock
About the team
The AWS Mantle team is building the next-generation inference engine that powers Amazon Bedrock—providing secure, enterprise-grade access to high-performing foundation models from the world's leading AI companies. Our mission is to simplify and accelerate how models are served at global scale, with an unwavering commitment to customer trust through innovations like our Zero Operator Access architecture, designed so that no person—whether from AWS, a customer, or a model provider—can ever access customer inference data.
- We operate at massive scale, serving inference requests across all major AWS regions with sophisticated automated capacity management and unified resource pools
- Our team values builders who thrive in ambiguity, think long-term, and are excited to define the future of AI infrastructure from the ground up
- We foster a collaborative, inclusive environment where diverse perspectives drive better solutions—and where the best ideas win regardless of where they originate
- We ship fast and iterate with purpose, having rapidly expanded from launch to supporting models from OpenAI, DeepSeek, Google, Mistral, NVIDIA, and more
- We believe work should be meaningful and fun—you'll join a team that takes pride in making history at the forefront of generative AI
- Define and own very large programs spanning multiple engineering teams, product organizations, and VP orgs—driving cross-functional alignment on Mantle's most critical technical initiatives
- Operate in deep ambiguity where the business and architectural strategy may not yet be defined; drive clarity and create the path forward
- Identify risks and opportunities in technical strategies, architectures, and engineering organization structures, influencing decision-makers at the VP level
- Make strategic trade-offs between business opportunity, resource allocation, and system sustainability while remediating critical and endemic problems
- Effectively force-multiply across the Mantle organization—accelerating delivery, managing escalations, and aligning teams toward simple, coherent approaches
Key job responsibilities
As a Principal TPM on the Mantle team, you will be the strategic connective tissue across our distributed inference platform—orchestrating the delivery of programs that span model onboarding, capacity management, security architecture, and global API availability. You will not simply manage timelines; you will define missions, set objectives, analyze data, and influence resource allocation at the highest levels.
- Own end-to-end program strategy for Mantle's most complex cross-organizational initiatives, including global model onboarding pipelines, Zero Operator Access security programs, and multi-region capacity scaling
- Drive technical architecture discussions with Principal Engineers and SDMs, identifying dependencies, scaling factors, boundary conditions, and risks across distributed systems serving ML inference at global scale
- Represent and advocate for customers during executive-level prioritization and planning, owning roadmap strategy and ensuring programs deliver measurable outcomes
- Create and optimize cross-org structures and mechanisms that align teams toward coherent delivery approaches—eliminating unnecessary process and streamlining execution
- Actively develop the TPM community through mentorship, promotion assessments, and raising the bar on program management excellence across Amazon Bedrock
About the team
The AWS Mantle team is building the next-generation inference engine that powers Amazon Bedrock—providing secure, enterprise-grade access to high-performing foundation models from the world's leading AI companies. Our mission is to simplify and accelerate how models are served at global scale, with an unwavering commitment to customer trust through innovations like our Zero Operator Access architecture, designed so that no person—whether from AWS, a customer, or a model provider—can ever access customer inference data.
- We operate at massive scale, serving inference requests across all major AWS regions with sophisticated automated capacity management and unified resource pools
- Our team values builders who thrive in ambiguity, think long-term, and are excited to define the future of AI infrastructure from the ground up
- We foster a collaborative, inclusive environment where diverse perspectives drive better solutions—and where the best ideas win regardless of where they originate
- We ship fast and iterate with purpose, having rapidly expanded from launch to supporting models from OpenAI, DeepSeek, Google, Mistral, NVIDIA, and more
- We believe work should be meaningful and fun—you'll join a team that takes pride in making history at the forefront of generative AI