Senior Site Reliability Engineer — Autheo | Live Mainnet, RPC & Validator Operations

Launch Legends, together with Autheo, is building enterprise-grade blockchain and decentralized infrastructure solutions. The company is seeking a Senior Site Reliability Engineer to operate the live Autheo Mainnet’s validator, sentry, RPC, cloud, observability, security, and incident-response infrastructure while supporting production bridge, API, Marketplace, TheoID, and THEO AI surfaces.

Responsibilities:

• Operate validator, sentry, and RPC infrastructure for Autheo Mainnet. Full and archive where the product needs them
• Watch the chain, not a generic host: block height, peers, sync, missed blocks, mempool, and endpoint error rate
• Capacity for public RPC: load balancing, caching, throttling, and a clear split between public and private peers
• Production on AWS: VPC, IAM, KMS, Secrets Manager, load balancers, and the compute the nodes actually run on
• Terraform for anything you expect to rebuild. Kubernetes (EKS or equivalent) where the service is a service, not a node that must not move
• Helm or GitOps for the deploy path. A console click is not a change
• SLOs for the paths that matter: RPC availability, bridge, DevHub API, staking and fee submission. Error budget in writing
• On-call for S1 and S2. Incident command, a timeline, and a review that names the next control
• Toil removed by automation. CI for Terraform plan, image build, and Helm. Python, Bash, or Go
• Metrics, logs, and traces on the paths you own. Prometheus, Grafana, OpenTelemetry, CloudWatch. Datadog or ELK if that is what is already running
• Alerts a human can act on. A page without a runbook is a draft
• KMS and secret handling for node keys, bridge keys, and service credentials. No key in a repo, a chat, or a disk image you cannot account for
• Guardrails that match the audits already done (Halborn on the validator platform, CertiK on Mainnet, Adevar on the Base bridge). SOC 2 work only where a control owner exists
• A map of what is in production versus what is still a laptop: validators, RPC, bridge, DevHub, and the rollout surfaces
• One SLO with an error budget, and the dashboard that shows it
• On-call that has a roster, a severity definition, and a runbook for RPC stall and validator miss
• Terraform or Helm for one path that today is a manual rebuild
• One repeat incident turned into a control, with the review written down

Requirements:

• 7+ years in cloud, SRE, systems, or DevOps, including production on-call
• 5+ years operating production workloads on AWS
• Kubernetes and containers in production, and Terraform you have applied, not only planned
• Observability you can walk through: Prometheus, Grafana, OpenTelemetry, CloudWatch, or Datadog
• A blockchain network you have run or supported: nodes, RPC, sync, and a failed transaction you traced to a cause
• Distributed systems: retries, backpressure, failover, and a capacity number you measured
• Incident response and a post-incident review you can share
• Cosmos SDK or EVM client operations, IBC relayers, or a bridge you have kept up
• GitOps (Argo or equivalent) and a pipeline you owned
• AWS certification. Useful, not required
• A runbook or a public postmortem