HERC RENTALS INTERVIEW PREP Cloud and Platform Engineering Manager WHAT WE LEARNED IN ROUND ONE

HERC RENTALS INTERVIEW PREP
Cloud and Platform Engineering Manager

WHAT WE LEARNED IN ROUND ONE

Team is 5 engineers, all senior and technically strong.
Main problem is no clear structure or daily plan, people work in different directions.
App teams do not fully understand the pipelines and need a simple standard path.
Architecture and infra decisions are made together as a team, but there is no formal process.
They use EKS for many microservices and ECS Fargate for small apps.
Identity is Okta.
AI is still at concept stage, integrating into backend apps, user access is limited, and they are careful because people misuse it.
The manager wants someone who observes the current process first, finds gaps day to day, works close with another manager, and removes leadership blockers.
Job post clouds are AWS, Azure, OCI with Terraform, CloudFormation and Bicep.
Job is on-site in Bonita Springs, Florida.

YOUR PITCH, TELL ME ABOUT YOURSELF

Thank you for having me again. I am a DevOps engineer with more than five years of experience on AWS. Four years at Yelp and now at Zulily. My daily work is Terraform, Kubernetes on EKS, GitHub Actions pipelines and CloudWatch monitoring. At Yelp I improved deployment speed by 35 percent with blue-green deployments and automatic rollback. I also mentored junior engineers, wrote runbooks and helped onboard new people, and that is the part I enjoy most. In our first conversation I learned your team is very senior but needs structure and a clear path for app teams. That is exactly what I like to build. I want to stay hands-on and also help the team move in one direction.

Numbers to remember: 5 plus years, 35 percent faster deployments, CKAD January 2025, AWS Solutions Architect Associate August 2026.

LEADERSHIP ANSWERS

What is your leadership experience?
I have not been a formal manager with direct reports yet. But at Yelp I mentored junior engineers, onboarded new people, wrote runbooks and led knowledge sharing. A team of five is a good size for me, I can coach and still be hands-on.

How would you lead senior engineers?
I would not micromanage. I give clear goals and priorities, and each senior person owns one area, like Terraform, CI/CD or EKS. I stay hands-on with them on hard problems. Senior people want trust and direction, not control.

How do you resolve conflict between two engineers?
I bring both together and let each explain without interrupting. We go back to the goal, cost, security, reliability or time. I ask for facts, data or a small proof of concept. If they still disagree, I decide, explain why, and write it down. Then we disagree and commit. At Yelp two engineers disagreed on speed versus safety, and we chose blue-green with automatic rollback, which gave us 35 percent faster deployments.

One engineer is not meeting expectations, what do you do?
Private one on one first, I listen and ask what is going on. Then I make expectations clear with examples and small goals with dates. I give support, pair them with a strong engineer and remove blockers. Weekly check-ins. If no progress, I work with HR on a fair, documented plan.

How do you prioritize when everything is urgent?
Business impact and risk. Production and security issues first, then work that unblocks app teams, then improvements. I agree on top priorities with my manager every week so the team is not pulled in many directions.

How do you report to leadership?
Short and simple. A monthly update with roadmap, risks, cloud cost trend, and a few metrics like deployment frequency, failed changes, incidents and time to recover.

How do you handle an outage as the escalation point?
First stop the impact with rollback or failover. Update people every 30 minutes. One person leads, others fix. After, a blameless postmortem with clear action items and owners.

TECHNICAL ANSWERS FROM ROUND ONE

Recent architecture you deployed?
At Zulily, an app on EKS with DocumentDB. VPC across multiple availability zones. Public subnets only for the load balancer, private subnets for EKS nodes and the database. Route 53 to the load balancer with TLS, then to pods. DocumentDB is private, only the app security group can reach it, IAM least privilege. HPA adds pods, Cluster Autoscaler adds nodes, so it grows during sales and shrinks after. Everything in Terraform modules, same code for dev, staging and prod. GitHub Actions with tests, terraform plan review and approval before prod. CloudWatch for alerts.

Kubernetes experience?
Four years on EKS at Yelp and now Zulily, CKAD certified. Helm charts so every team deploys the same way. Liveness and readiness probes for self-healing. HPA and Cluster Autoscaler. I troubleshoot CrashLoopBackOff, image pull errors and resource limits with kubectl describe, logs and events.

EKS versus ECS Fargate?
EKS fits many microservices, more control, Helm, one deployment pattern. Fargate fits small apps, no nodes to manage, less work. My hands-on is EKS, ECS is new for me but the ideas are very close.

Event based architecture and Lambda?
Services do not call each other directly, one sends an event and others react. EventBridge routes events, SNS sends to many, SQS queues so nothing is lost, Lambda runs code when an event comes. Watch the 15 minute timeout, memory, cold starts, IAM role, dead letter queues, and make processing safe to run twice. I am familiar, but have not built big systems myself.

Identity?
AWS IAM roles and least privilege. GitHub Actions deploys with IAM roles, no long term keys. IAM roles for Kubernetes service accounts on EKS. Okta and Active Directory I know the ideas, single sign on, MFA, groups mapped to AWS roles through IAM Identity Center.

How do you use AI?
Claude AI daily as a helper. Draft pipelines, review Terraform, explain errors, write documentation. Always review and test, never share secrets or customer data. I agree with your careful approach, start small, approved tools, human review, and guardrails like logging and access control.

NEW TOPICS LIKELY IN ROUND TWO

Landing zones
Key words: AWS Organizations, Control Tower, separate accounts per environment, SCPs, central logging.
Answer: A landing zone is the secure starting setup for all cloud accounts. Separate accounts for prod, non-prod, security and logging, with guardrails so teams cannot break the rules.

Azure basics
Key words: subscriptions, resource groups, VNet, AKS, Entra ID, Bicep.
Answer: Azure is very similar to AWS. VNet is like VPC, AKS is like EKS, Entra ID is like IAM, Bicep is like CloudFormation. Terraform works on both.

OCI basics
Key words: tenancy, compartments, VCN, OKE.
Answer: OCI uses compartments to organize resources, VCN is like VPC, OKE is their Kubernetes. Often used for Oracle databases and ERP.

Terraform at scale
Key words: modules, remote state, state locking, separate folders per environment, plan review in pull request.
Answer: I already do this at Zulily with S3 state and DynamoDB locking. The same pattern works with Azure and OCI providers.

Hybrid networking
Key words: site to site VPN, Direct Connect, ExpressRoute, Transit Gateway, private DNS.
Answer: It connects on-prem to cloud. VPN is quick and cheap, Direct Connect or ExpressRoute is private and faster for production.

Disaster recovery
Key words: RTO, RPO, backup and restore, pilot light, warm standby, multi-region.
Answer: RTO is how fast we recover, RPO is how much data we can lose. We choose the strategy by business need and cost, and we test restores regularly.

Cloud cost, FinOps
Key words: tagging, right-sizing, Savings Plans, Reserved Instances, spot, Cost Explorer, budgets and alerts.
Answer: First tag everything so we know who spends what. Then right-size, turn off unused dev at night, Savings Plans for steady work, spot for flexible work. Monthly cost report to leadership.

Security
Key words: Secrets Manager, KMS encryption, GuardDuty, Security Hub, image scanning, CIS benchmarks.
Answer: Least privilege, encrypt everything, no secrets in code, scan images in the pipeline, and work with the security team to fix findings by priority.

Observability
Key words: metrics, logs, traces, CloudWatch, Datadog, Prometheus and Grafana, alert fatigue.
Answer: Alert only on what users feel, every alert has a runbook, same dashboards across clouds.

Self-service platform
Key words: golden path, pipeline templates, Helm and Terraform modules, developer portal like Backstage.
Answer: App teams get a standard template and can deploy alone, the platform team keeps guardrails. This fixes the pipeline confusion they told us about.

Vendors and managed services
Key words: SLAs, quarterly reviews, cost, escalation path.
Answer: Clear SLAs, regular review meetings, and our team still understands what the vendor runs.

HONEST GAP ANSWERS

Pattern for every gap: say honestly you have not done it in production, show you know the idea, connect it to something you did, end with how fast you learn.

Formal manager: Not formal yet, but mentoring, onboarding, runbooks and knowledge sharing at Yelp. Ready for the next step.
Azure and OCI: Not in production, my work is AWS. Concepts map one to one and Terraform works on all three, so I can be productive fast.
Bicep and CloudFormation: Terraform is my main tool. Same idea, declare resources, review the plan, deploy.
ECS: Hands-on is EKS. ECS task definitions and services are close to Kubernetes deployments.
Active Directory and Okta: Not managed them, but I know single sign on, MFA, groups mapped to cloud roles. Strong IAM background.
Event driven and Lambda at scale: Familiar, not built big systems. I can support it with Terraform, IAM and monitoring.
FinOps and DR ownership: Have not owned it fully, familiar with right-sizing, Savings Plans, backup and restore testing, RTO and RPO.
Vendor management: Not owned vendor contracts yet, but I know the basics, clear SLAs and regular reviews.

30 60 90 DAY PLAN

First 30 days, listen and learn. One on one with each of the 5 engineers and the other manager. Sit with app teams and watch where they get stuck with pipelines. Review the AWS, Azure and OCI setup, Terraform code, incidents and cloud cost. Write a gap list and agree on top priorities with my manager.

Days 31 to 60, quick wins and structure. Weekly planning and a short daily sync so work is visible. Each senior engineer owns one area. Simple design doc and weekly design review for bigger decisions. First standard pipeline template for app teams.

Days 61 to 90, scale it. Golden path documentation and runbooks so app teams onboard alone. Monthly report to leadership with roadmap, risks, cost and delivery metrics. Plan for cost tagging, DR testing and safe AI tools.

QUESTIONS TO ASK

Do not repeat what they already answered: team size, seniority, Okta, EKS and ECS, AI stage.

For the hiring manager or director:
What would make you say after 6 months, this was the right hire?
How is work split between AWS, Azure and OCI, and is there a plan to consolidate?
How does this role divide work with the other manager?
What is the biggest risk on the platform today?

For engineers on the team:
What slows you down most in a normal week?
How do app teams ask you for help today, tickets, Slack or meetings?
What would you want from a new manager?

For HR or recruiter:
Is there relocation support for the move to Bonita Springs?
Is there a leadership program for new managers?
What are the next steps and timeline?

LOGISTICS AND FINAL TIPS

Relocation: have a clear yes and a realistic start date ready, because you live in Vancouver, WA and the job is on-site in Florida.
Salary: research the range for cloud engineering managers in Southwest Florida and give a range, not one number.
About Herc: they rent equipment to construction, industrial and event customers. Read their latest news the day before and mention one recent fact.
Every answer: short, one real example, end with a question.
Use their words back: structure, observe first, golden path for app teams, careful with AI. It shows you listened.
Send a thank you email the same day and mention one thing you discussed.
Practice the pitch out loud three times.
Practice the Azure and OCI mapping to AWS.
Prepare one outage story and one cost saving idea.