Hi P***p, I’m J***e, an HR recruiter at StrategyBrain. I came across your profile and was impressed by your work as Managing Director at V***N,especially your leadership in API integrations, CRM optimization, and operational excellence across insurance processes. We’re partnering with I***H to hire a Director of Insurance Technology and Operations. In this role, you’d drive end-to-end digital solutions, mentor cross-functional teams, and align technology initiatives with strategic business goals. Would you be open to a brief chat to explore this opportunity? Looking forward to your thoughts. J***e

Hi P***p, Great to hear you’re interested! Here’s a bit more on the opportunity: I***H is a 150-person firm dedicated to transforming the insurance sector with cutting-edge API and CRM platforms. As Director of Insurance Technology and Operations, you would: • Lead end-to-end delivery of digital solutions for brokers and carriers , from CRM enhancements to secure API integrations across underwriting, policy admin and claims • Drive operational excellence by analyzing processes, defining KPIs and implementing continuous improvements • Build and mentor cross-functional teams (product, engineering, analysts), fostering an innovative, accountable culture • Partner with stakeholders to set the product roadmap, prioritize initiatives and ensure alignment with strategic objectives • Leverage data-driven insights to refine solutions and deliver measurable outcomes Why this fits your background: • You’ll apply your V***N experience in API integrations and CRM optimization at a broader, strategic level • You’ll gain full P&L and program ownership, influencing the roadmap for a growing suite of InsurTech products • You’ll have access to learning budgets, certifications and a collaborative environment that prizes fresh ideas We offer a highly competitive salary, bonus structure and stock-option plan to recognize your leadership impact. Does this resonate with what you’re looking for? I’d love to dive deeper on a quick 20-minute call. Are you available next Tuesday or Thursday morning CET? Let me know what works best, or feel free to suggest another time. Looking forward to your thoughts! Best, J***e HR Recruiter, StrategyBrain

Hi P***p, Happy to dive in. Currently, the Insurance Tech & Ops organization at I***H comprises about 12 people split into two pods: • Pod 1 (CRM & Customer Workflows): Led by a Senior Product Manager, with 3 software engineers, a UX specialist and a business analyst • Pod 2 (API & Integrations): Headed by an API Lead, with 2 backend engineers, a QA engineer and another analyst • You’d partner closely with operations (claims, underwriting, policy admin) and our customer-success team, plus external implementation partners. Key challenges you’d tackle day one: 1. Scaling our API platform to support high-volume, low-latency insurer connections 2. Harmonizing disparate underwriting and claims processes into a single, data-driven workflow 3. Defining and tracking KPIs that align both technology and business stakeholders 4. Driving adoption of new CRM features across brokers with varying tech maturity Thursday at 10:00 AM CET works for me,shall I send a calendar invite with a Teams link? Looking forward to our chat. Best, J***e HR Recruiter, StrategyBrain

Hi P***p, Great,I’ve just sent the Teams invite for Thursday at 10:00 AM CET. As for our API platform, we run a microservices architecture primarily built in Java Spring Boot (with a few Node.js services) deployed on Kubernetes in AWS. We expose REST endpoints (some GraphQL) secured via OAuth2 and manage them through AWS API Gateway. For monitoring, we use Datadog to track key metrics. Currently our baseline throughput sits in the low hundreds of requests per second during peak, with average end-to-end latency around 150,200 ms and p95 latency under 350 ms. We’ll have our API Lead on the call to dive deeper into SLAs, performance tuning and upcoming enhancements. Looking forward to discussing more on Thursday! Best, J***e HR Recruiter, StrategyBrain

Hi P***p, Great question,here’s a high-level view of how we handle autoscaling and CI/CD for performance: 1. Autoscaling in AWS - Pod-level: Kubernetes Horizontal Pod Autoscaler (HPA) driven by Datadog-sourced metrics (CPU, memory, request-per-second and custom business metrics like queue length or error rate). - Cluster-level: AWS Cluster Autoscaler adjusts our EKS node groups (a mix of on-demand and spot instances) to ensure capacity. We also use scheduled and predictive scaling for known traffic peaks. 2. CI/CD & Performance Testing - Pipeline triggers on Git merges: Docker images are built, security-scanned, and undergo unit & integration tests. - Staging deployment via Helm charts: automated smoke tests, end-to-end API tests and lightweight load tests run here. - Canary/Blue-Green in production: we roll out performance optimizations in small slices, monitor with Datadog APM and logs, and auto-rollback on anomalies. - Custom load testing (e.g. JMeter or Gatling scripts) runs in a pre-prod environment to validate SLAs before full rollout. This setup ensures we catch regressions early, scale seamlessly under load, and have the data to refine optimizations continuously. It aligns closely with your V***N experience,our API Lead can dive deeper on Thursday’s call. Let me know if you’d like any further details ahead of 10 AM CET. Looking forward to our conversation! J***e

Hi P***p, Great questions,here’s a bit more detail: 1. Custom Metrics for HPA • Queue/backlog length (e.g. pending requests in our internal work queues) • 4xx/5xx error rates (rolling 1- or 5-minute windows) • End-to-end latency p95 (via Datadog APM) • Business-specific counters (active sessions, in-flight transactions) These metrics flow into Kubernetes through the Datadog metrics adapter, driving our HPA alongside CPU and memory. 2. Spot vs. On-Demand Node Balancing • We maintain a baseline of on-demand nodes (≈25,30% of capacity) to guarantee headroom. • Spot instances cover the bulk of burst capacity (60,70%), spun up by the AWS Cluster Autoscaler. • A mixed-instance ASG with fallback to on-demand ensures we never lose scaling headroom if spot capacity dips. • For predictable spikes, we also lean on scheduled scaling policies to pre-warm extra nodes. 3. Canary / Blue-Green Rollout Gates • We start with a 5,10% traffic slice for 5,10 minutes. • Auto-rollback triggers if: , Error rate exceeds 1% over two consecutive intervals , p95 latency deteriorates >10% vs. baseline • Synthetic end-point checks must remain ≥99% healthy. • After canary passes, we do a quick manual review before cutting over 100%. Happy to let our API Lead and SRE walk through any of these thresholds in more depth on Thursday. Let me know if there’s a specific metric or scenario you’d like us to prepare ahead of our session. Best, J***e

Hi P***p, Great questions ahead of our chat,here’s a bit more color: 1. Queue/Backlog Metric • We built a small custom Datadog exporter in Python that pulls from our internal work-queue API and pushes metrics via the Datadog Agent. • From there we leverage the official Datadog metrics adapter into Kubernetes so HPA can scale on these business-level signals. • We evaluated KEDA but chose the custom exporter to keep all our metrics centralized in Datadog and support richer tagging. 2. Spot Reclaim & Pod Eviction • We run the AWS Node Termination Handler (cordon/drain mode) as a DaemonSet, which captures the two-minute spot termination notice. • A preStop hook on our critical pods runs a brief drain script that respects PodDisruptionBudgets and gracefully finishes in-flight requests. • During that window, Cluster Autoscaler backfills capacity with on-demand nodes if spot capacity dips, ensuring no SLA impact. Happy to have our SRE and API Lead drill into any of this on Thursday at 10 AM CET. Let me know if there’s anything else you’d like us to prep! Looking forward to our conversation, J***e