Senior Security Solution Architect, APJC Customer Security, APJC Customer Security
Amazon · Seoul, KOR ·
- Seniority
- Senior
- Category
- Architecture
Amazon Web Services (AWS) is seeking a Senior Secure AI Solution Architect to join our APJC Customer Security team. This is a builder's role. Customers across the region are moving fast on AI and seeking guidance on challenges they have not faced before — how to give autonomous agents an identity, how to defend against prompt injection, how to govern the AI development lifecycle, and how to produce the governance evidence a regulator will trust. You are the architect who helps them get there, working hands-on in their environment to build the prototypes and proofs of concept that turn ambition into validated proof — often in days.
You operate at the frontier, where the path forward is still being defined. Customers have a clear objective: put agentic AI into production while managing risk, regulatory, and data-residency obligations. Your role is to design the way there — creating new approaches where no established pattern exists and applying judgment about each customer's operational context and regulatory environment. You are technically self-sufficient: you write the code, deploy it with IaC, harden it, and demonstrate it live against real data.
Depth alone is not the job — scale is. The hardest problems you solve with one customer should lift the whole field. You independently dive deep on complex, novel Secure AI challenges, produce original technical content (reference architectures, working code, patterns, workshops), and share it so a pattern proven with one customer becomes something every area can reuse. You own the relationship with AWS service teams (Bedrock, AgentCore, SageMaker) — you are the voice of the field that shapes the product roadmap, and the voice of the product that shapes field guidance.
Security and governance are what let customers say yes to AI with confidence. You give the CISO the frameworks, evidence, and tooling to move quickly and safely, and you give the CIO/CTO confidence that governance won't slow delivery. You are the proof that AWS is the best place to build securely with AI.
Key job responsibilities
Key job responsibilities
Build working prototypes and proofs of concept in customer sandbox environments — using AWS AI services (Bedrock, AgentCore, SageMaker), AI-augmented development tooling, and IaC (CDK, Terraform) — that prove a Secure AI control works against the customer's data and risk model, in days rather than quarters.
Implement directly to unblock decisions: stand up agentic-identity, prompt-injection-defence, AI development lifecycle, guardrail, and audit/evidence patterns that validate what conversations alone cannot.
Stay engaged through implementation so the design holds against reality; exercise judgment on when to revise the approach based on what you learn.
Partner with the Bedrock, AgentCore, and SageMaker service teams — you are the voice of APJC field security requirements that shapes what ships, and the voice of the product that shapes what the field advises.
Feed customer voices back through feature requests and direct engagement, staying current with a threat and regulatory landscape that moves in hours.
Translate service-team roadmap into field-ready guidance, ensuring it gets packaged at scale.
Produce original technical content — reference architectures, working code, patterns, blog posts, and workshops — that peers and customers adopt, turning a single customer win into reusable regional IP.
Mentor and develop others — area-based Solutions Architects and the wider field community — raising the Secure AI technical bar across the region and growing the next set of specialists.
Hand proven patterns to the content and enablement team so guidance reaches the field without requiring your presence.
A day in the life
You start the morning in a customer's sandbox, not a slide deck. A financial-services customer wants to launch an agentic application on Bedrock, and overnight the identity and audit controls you built produced exactly the evidence their CISO needed. You tidy the code, capture the pattern, and prep a short summary naming the two guardrail settings that gave them confidence to proceed.
Mid-morning you join your Security Sales partner and the area account team with a manufacturer who wants to build with Bedrock and needs guidance on governing data residency and PII. It is a genuinely new problem — no established pattern exists — so you design the way forward: you whiteboard an agentic-identity and data-residency approach, then commit to prototyping the non-human-identity pattern their builder team will adopt.
After lunch you go heads-down and build it — hands on keyboard, deploying to a sandbox, hardening it — then run a working session with a customer's platform team on prompt-injection defence for a coding-assistant rollout.
You close the day scaling beyond yourself: you package this week's FSI win into a reference architecture the other areas can reuse, hand it to the content team to turn into a workshop, share a write-up with the field, and file a feature request for the AgentCore team based on what you learned. No two days look the same, but the thread is consistent: you help a customer accelerate securely, then turn what you built into something the whole region benefits from.
About the team
APJC Customer Security helps customers across the region adopt AI on AWS with confidence — ensuring security, risk, and regulation accelerate AI adoption rather than slow it down. We run two coupled programmes: Secure AI (making AWS the best place to build securely with AI) and AI for Security (using AI to help customers defend themselves). As a Senior Secure AI Solution Architect, you are the deep technical expertise brought directly to where customers need it. You define the approach on novel problems, prove it hands-on, then hand the pattern to the content team and the field to scale. You will have a clear development path, a peer community of specialists, direct leadership visibility, and the autonomy to set technical direction in one of the most consequential areas in the industry.
You operate at the frontier, where the path forward is still being defined. Customers have a clear objective: put agentic AI into production while managing risk, regulatory, and data-residency obligations. Your role is to design the way there — creating new approaches where no established pattern exists and applying judgment about each customer's operational context and regulatory environment. You are technically self-sufficient: you write the code, deploy it with IaC, harden it, and demonstrate it live against real data.
Depth alone is not the job — scale is. The hardest problems you solve with one customer should lift the whole field. You independently dive deep on complex, novel Secure AI challenges, produce original technical content (reference architectures, working code, patterns, workshops), and share it so a pattern proven with one customer becomes something every area can reuse. You own the relationship with AWS service teams (Bedrock, AgentCore, SageMaker) — you are the voice of the field that shapes the product roadmap, and the voice of the product that shapes field guidance.
Security and governance are what let customers say yes to AI with confidence. You give the CISO the frameworks, evidence, and tooling to move quickly and safely, and you give the CIO/CTO confidence that governance won't slow delivery. You are the proof that AWS is the best place to build securely with AI.
Key job responsibilities
Key job responsibilities
Build working prototypes and proofs of concept in customer sandbox environments — using AWS AI services (Bedrock, AgentCore, SageMaker), AI-augmented development tooling, and IaC (CDK, Terraform) — that prove a Secure AI control works against the customer's data and risk model, in days rather than quarters.
Implement directly to unblock decisions: stand up agentic-identity, prompt-injection-defence, AI development lifecycle, guardrail, and audit/evidence patterns that validate what conversations alone cannot.
Stay engaged through implementation so the design holds against reality; exercise judgment on when to revise the approach based on what you learn.
Partner with the Bedrock, AgentCore, and SageMaker service teams — you are the voice of APJC field security requirements that shapes what ships, and the voice of the product that shapes what the field advises.
Feed customer voices back through feature requests and direct engagement, staying current with a threat and regulatory landscape that moves in hours.
Translate service-team roadmap into field-ready guidance, ensuring it gets packaged at scale.
Produce original technical content — reference architectures, working code, patterns, blog posts, and workshops — that peers and customers adopt, turning a single customer win into reusable regional IP.
Mentor and develop others — area-based Solutions Architects and the wider field community — raising the Secure AI technical bar across the region and growing the next set of specialists.
Hand proven patterns to the content and enablement team so guidance reaches the field without requiring your presence.
A day in the life
You start the morning in a customer's sandbox, not a slide deck. A financial-services customer wants to launch an agentic application on Bedrock, and overnight the identity and audit controls you built produced exactly the evidence their CISO needed. You tidy the code, capture the pattern, and prep a short summary naming the two guardrail settings that gave them confidence to proceed.
Mid-morning you join your Security Sales partner and the area account team with a manufacturer who wants to build with Bedrock and needs guidance on governing data residency and PII. It is a genuinely new problem — no established pattern exists — so you design the way forward: you whiteboard an agentic-identity and data-residency approach, then commit to prototyping the non-human-identity pattern their builder team will adopt.
After lunch you go heads-down and build it — hands on keyboard, deploying to a sandbox, hardening it — then run a working session with a customer's platform team on prompt-injection defence for a coding-assistant rollout.
You close the day scaling beyond yourself: you package this week's FSI win into a reference architecture the other areas can reuse, hand it to the content team to turn into a workshop, share a write-up with the field, and file a feature request for the AgentCore team based on what you learned. No two days look the same, but the thread is consistent: you help a customer accelerate securely, then turn what you built into something the whole region benefits from.
About the team
APJC Customer Security helps customers across the region adopt AI on AWS with confidence — ensuring security, risk, and regulation accelerate AI adoption rather than slow it down. We run two coupled programmes: Secure AI (making AWS the best place to build securely with AI) and AI for Security (using AI to help customers defend themselves). As a Senior Secure AI Solution Architect, you are the deep technical expertise brought directly to where customers need it. You define the approach on novel problems, prove it hands-on, then hand the pattern to the content team and the field to scale. You will have a clear development path, a peer community of specialists, direct leadership visibility, and the autonomy to set technical direction in one of the most consequential areas in the industry.