Software Engineer 2 Infra

About the team

The Core Infrastructure team builds the platforms and systems that power everything at Fam, from real-time app experiences to high-volume payments. It operates as a small, high-leverage team with deep ownership. Every service at Fam runs on the platform this team builds, so it cares obsessively about system fundamentals, operational excellence, and mechanical sympathy.

What our systems process

• 1 Billion+ API requests per day powering ~10M daily transactions for 10M+ users at 20,000+ RPS.
• 470+ node Kubernetes cluster (~2,500 vCPUs and ~8 TiB RAM) under active orchestration.
• Dynamic, multi-dimensional auto-scaling powered by 195+ HPAs paired with Karpenter.
• 12+ TB of active transactional data (and counting) across our high-throughput database clusters.
• Event Streaming - 400+ topics, and 3,800+ partitions at ~46,700 messages/sec

Our Next Chapter

While our core systems already handle massive scale, our next vision as a team is to turn infrastructure into a frictionless platform discipline.

We are building towards a world where internal engineering teams consume platform primitives compute, storage, and event streams via self-serve contracts without waiting on us. We're also laying the groundwork for AI-native platform operations, developing agent plane and agent skills so agents can safely interact & operate infrastructure primitives alongside our team. We look to the CNCF landscape for best-in-class tools our systems, with a long-term commitment to contributing back to open source.

What You'll Do

Build and operate compute, storage, and networking at scale that powers FamApp.
Design platform contracts for other teams: Modules, golden paths, and abstractions with clear inputs, outputs, and guarantees.
Take medium-to-large infrastructure projects from design to rollout, manage dependencies across engineering teams, and work through ambiguity without waiting to be unblocked.
Build idempotent, retry-safe automation with its own observability, integrated with CI/CD, ticketing, compliance, and incident workflows.
Build reusable infrastructure with Infrastructure as Code and Configuration as Code with clear interfaces, validation and versioning backed by safe provisioning ci/cd workflows.
Operate Kubernetes as the Operating system - Own the cluster lifecycle (worker plane, node groups, add-ons) . Ability to extend it with controllers, operators, and CRDs so teams consume platform capabilities as Kubernetes-native APIs.
Define user-facing SLIs, SLO targets, burn-rate alerts, and error-budget actions for what you own. Carry the pager, cut alert noise, and lead post-incident follow-up with blameless postmortem.
Track cloud spend for your services and look for savings without compromising reliability or compliance.
Maintain golden pipeline templates, GitOps release patterns, and self-service onboarding, with security gates and least-privilege pipeline identities built in, which enables developers to ship their code to different environments.
Turn requirements into technical design docs that cover the full life of a system: how teams adopt it, how it's operated and maintained, and how it stays reliable (monitoring, SLIs, backup, DR, incident response), so anyone on the team can follow them.

Must Haves

3-6 years in DevOps, SRE, or platform engineering, with ownership of production systems.
Cloud compute and Linux fundamentals - Instances, images, and block storage on any major cloud: diagnosing throughput/IOPS limits, memory pressure, inode and file-descriptor exhaustion, and systemd issues, and correlating OS-level evidence with cloud metrics and dashboards.
Cloud networking - Network and subnet design, CIDR planning, routing, firewall and security-group rules, and DNS. You can design cross-account or cross-project connectivity, weigh peering against transit hubs and private endpoints, and debug asymmetric routing and overlapping CIDRs.
Hands-on Kubernetes in production managed services like EKS, GKE, or AKS.
Terraform or equivalent IaC with reusable, versioned modules and safe change workflows.
Programming ability in Python or Go to build automation that is idempotent, retry-safe, and observable.
CI/CD and GitOps experience - pipeline templates, release and rollback patterns, security gates.
Reliability practice: SLIs/SLOs, actionable alerting, on-call, and blameless post-mortems.
Everyday use of AI tools, with the habit of validating their output before it ships.

Good to Have

Karpenter in production - consolidation, spot handling, multi-NodePool strategy.
Observability at scale: Prometheus/VictoriaMetrics, Loki, Jaeger, OpenTelemetry, high-cardinality metrics.
CI/CD supply-chain security (SAST, secret detection, image scanning, artifact promotion).
ClickHouse Operator on kubernetes or another analytical store solutions managed at scale in production cluster.
PCI, RBI, SOC 2, or similar compliance-heavy environments.
Experience in building scaling and managing infrastructure for Customer focused products.