Teams at Appnovation

Own reliability and operations for a self-hosted LLM observability platform running on Kubernetes and AWS. Responsibilities include operating ClickHouse, PostgreSQL, Redis, and ingestion workers; implementing monitoring, alerting, and service levels; testing backups and restores; safely upgrading software; maintaining infrastructure as code; automating recurring tasks; and creating runbooks, SOPs, onboarding, and support processes.

Build and operate reliability and observability capabilities for an internal AWS Bedrock AgentCore platform. Responsibilities include OpenTelemetry instrumentation, CloudWatch and enterprise observability integrations, distributed tracing, SLOs, alerting, token and compute cost monitoring, dashboards, runbooks, incident response, and post-incident improvements. The role requires strong AWS, SRE, observability, FinOps, scripting, and coding experience, with preferred exposure to LLM monitoring, infrastructure as code, and regulated industries.

Build the agent-facing platform layer, including runtime patterns, memory, orchestration, multi-agent handoffs, governed tools, MCP servers, authentication, legacy integrations, evaluations, and guardrails. Develop production-quality reference implementations and standards using AWS and agent frameworks. Support enterprise teams through documentation, walkthroughs, and enablement while ensuring secure access, reliable error handling, testing, and reusable code.