The AI Compute Centre within Tencent's Overseas IT department is seeking a Senior AI Infrastructure Engineer to bridge the gap between internal AI infrastructure demand and external resource fulfillment. This role involves leading cloud and edge cloud infrastructure projects across global regions, managing the full lifecycle of AI infrastructure clusters from requirement gathering to operational handover. Responsibilities include translating business demands into technical infrastructure requirements, leading network design reviews, managing network providers and delivery dependencies, and ensuring data center readiness. You will oversee integration and acceptance governance, maintain project plans, and communicate technical risks to stakeholders. Requirements include 10+ years of experience in infrastructure architecture, network engineering, or technical delivery, with strong expertise in data center networking, TCP/IP, routing protocols (BGP, OSPF), and managing complex infrastructure transitions. Candidates should possess strong stakeholder communication skills and the ability to manage suppliers through design and delivery milestones. Preferred qualifications include experience with AI server capacity sourcing, high-performance networking (InfiniBand, RoCE, RDMA), high-density data center deployments, and relevant industry certifications.
For full details about this role, click “Apply on company site” to view the employer’s original posting.
Skills for this role
Infrastructure ArchitectureNetwork EngineeringData Center NetworkingTCP/IPL2/L3 NetworkingVLANsBGPOSPFCloud InfrastructureSupplier ManagementProject ManagementSpine-Leaf FabricsEVPN-VXLANInfiniBandRoCERDMANCCL
Your skill match
Checking your profile…
YOUR NEXT STEP
Get interview-ready for this role
A focused preparation guide, built around this job’s responsibilities and requirements.
✦ AI-generated guide
Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.
01
Understand the role
What matters most
End-to-End AI Infrastructure Lifecycle Management
The role requires managing the full lifecycle of AI clusters, from initial requirement gathering and capacity planning to operational handover. You must demonstrate the ability to translate abstract business demands into concrete technical specifications, supplier requirements, and acceptance criteria.
Network Design and Delivery Governance
You are responsible for leading network design reviews, including data center fabrics and WAN connectivity. You must ensure that designs meet high-performance AI workload requirements while maintaining day-to-day operability, and you must validate technical evidence for connectivity, latency, and failover.
Supplier and Stakeholder Coordination
Success depends on managing external vendors, carriers, and colocation partners to meet project milestones. You must be able to hold suppliers accountable for deliverables, resolve gaps between vendor commitments and actual outcomes, and communicate technical risks to internal business stakeholders.
Data Center Readiness and Integration
You must oversee the physical and logical readiness of data centers, including power, cooling, and rack space. This involves coordinating cross-functional teams (compute, storage, facilities) to ensure seamless integration and successful handover to operations.
Likely challenges to think through
Bridging Technical Gaps in AI Workloads
AI infrastructure often requires specialized networking like InfiniBand or RoCE. The challenge is that you may not be the primary engineer configuring these systems, yet you must be technically proficient enough to challenge assumptions and identify performance bottlenecks that could impact AI training or inference.
Managing Complex Global Dependencies
Delivering infrastructure across global regions involves coordinating multiple external providers with varying service levels. You should prepare to discuss how you manage conflicting timelines, resolve escalation issues between vendors, and maintain project momentum when faced with site-specific delays.
Defining Acceptance for Non-Standard Performance
Unlike traditional IT, AI clusters require rigorous validation of bandwidth and latency. The challenge is defining and enforcing acceptance criteria that go beyond simple 'up/down' status to ensure the cluster meets the specific performance outcomes required for AI workloads.
03
Practice your answers
Technical practice
Designing High-Performance AI Fabrics
What to coverDiscuss the trade-offs between InfiniBand and RoCE (RDMA over Converged Ethernet) for large-scale AI training clusters. Address how you would evaluate network topology choices like Spine-Leaf or Clos fabrics to minimize latency and maximize throughput for NCCL-based workloads.
Data Center Readiness and Power Constraints
What to coverExplain your approach to managing high-density data center deployments where power and cooling are limiting factors. Describe how you coordinate with facilities teams to ensure infrastructure readiness before hardware installation.
Network Resilience and Routing Protocols
What to coverHow do you ensure high availability in a multi-site data center environment using BGP and OSPF? Discuss strategies for managing network segmentation and redundancy to prevent single points of failure during large-scale infrastructure transitions.
Integration and Acceptance Governance
What to coverDescribe your methodology for validating that a newly deployed AI cluster meets performance requirements. Focus on how you move beyond basic connectivity checks to verify bandwidth, latency, and failover capabilities before operational handover.
Managing Complex Infrastructure Dependencies
What to coverWhen coordinating across compute, storage, and network teams, how do you identify and mitigate dependencies that could delay a cluster deployment? Provide an example of how you would handle a situation where a vendor's delivery timeline slips.
EVPN-VXLAN in Modern Data Centers
What to coverExplain the role of EVPN-VXLAN in modern data center networking and how it supports scalability and multi-tenancy. Discuss how you would review a design proposal to ensure it meets the specific needs of an AI compute environment.
Translating Business Demand to Technical Specs
What to coverHow do you translate high-level business requirements for AI capacity into specific technical infrastructure requirements? Suggest a framework for defining evaluation criteria and delivery milestones that align with stakeholder expectations.
Supplier and Service Accountability
What to coverHow do you hold external network providers and colocation partners accountable to SLAs? Discuss the metrics and documentation you would require to ensure service quality and effective incident escalation.
Behavioural practice
Managing Technical Conflict with Stakeholders
What to coverDescribe a situation where you had to communicate a significant technical risk or design trade-off to non-technical stakeholders. How did you ensure they understood the impact on the project timeline or business outcomes?
Driving Resolution in Cross-Functional Projects
What to coverProvide an example of a time you identified a gap between vendor commitments and actual delivery. What specific actions did you take to drive the issue to resolution without compromising the project's success?
Leading Infrastructure Transitions
What to coverShare an experience where you led a complex infrastructure transition, such as a data center migration or a large-scale deployment. How did you manage the coordination between multiple technical teams and external providers to ensure a smooth handover?
Challenging Assumptions in Design Reviews
What to coverDescribe a time when you had to challenge a technical assumption made by a specialist or vendor during a design review. What evidence did you use to support your position, and what was the final outcome?
02
Sharpen your knowledge
Topics to prepare
AI Infrastructure Lifecycle Management
Review the end-to-end process of cluster delivery, from requirement gathering and capacity planning to operational handover. Exercise: Draft a sample project plan that maps technical requirements (bandwidth, latency, power) to specific supplier delivery milestones and acceptance criteria. This is relevant to the role's focus on bridging internal demand with external resource fulfillment.
Data Center Network Architecture
Study modern high-performance data center fabrics, specifically Spine-Leaf topologies and inter-data-center connectivity. Exercise: Analyze a hypothetical network design for an AI cluster, identifying potential bottlenecks in redundancy or scalability. This is critical for the role's responsibility to lead network design reviews and ensure day-to-day operability.
Supplier and Service Governance
Understand how to manage vendor accountability through SLAs, technical risk logs, and acceptance testing. Exercise: Create a template for a 'Service Handover Document' that defines support boundaries, incident escalation paths, and monitoring expectations. This aligns with the requirement to manage providers and ensure operational readiness.
Familiarize yourself with the requirements for low-latency AI workloads, including RDMA over Converged Ethernet (RoCE) and InfiniBand. Exercise: Research the configuration prerequisites for NCCL performance testing in a multi-node GPU environment. This is a preferred qualification for managing high-performance AI infrastructure.
Deepen your knowledge of routing protocols and overlay technologies used in large-scale data centers. Exercise: Review a configuration scenario involving EVPN-VXLAN for multi-tenant network segmentation. This is essential for the role's requirement to review network topology and ensure robust connectivity.
Technical Stakeholder Communication
Practice translating complex infrastructure trade-offs into clear business impacts. Exercise: Prepare a brief summary of a hypothetical 'design defect' or 'capacity constraint' that explains the technical risk and proposed resolution to a non-technical stakeholder. This is vital for the role's cross-functional leadership responsibilities.
04
Your preparation checklist
Prepare a project case study on infrastructure delivery
Draft a structured walkthrough of a past large-scale data center deployment or infrastructure transition you led. Focus on how you managed cross-functional dependencies, resolved supplier gaps, and ensured successful operational handover.
Review AI-specific networking architectures
Refresh your knowledge on high-performance networking requirements for AI clusters, specifically Spine-Leaf fabrics, EVPN-VXLAN, and RDMA/RoCE/InfiniBand. Be prepared to discuss how these technologies impact latency and bandwidth for large-scale compute workloads.
Develop a framework for infrastructure acceptance governance
Create a sample checklist or governance plan for cluster-level acceptance. This should demonstrate how you validate connectivity, failover, and performance metrics against business requirements before signing off on a project.
Formulate questions on AI compute demand and supplier management
Prepare thoughtful questions regarding how the team currently balances internal AI demand with external provider SLAs, and how they handle technical risk communication when vendor delivery milestones are at risk.
Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy
Share this opportunityKnow someone who would be a great fit?