NET Observability

I'm a senior .NET backend developer looking to set up a full observability stack.

I know AI tools (Claude/Codex) can generate boilerplate — what I need is someone who's actually worked with these tools to collaborate as a peer developer.

We'll use AI tooling together to move fast, but your real-world experience is what makes this valuable.

You don't need to know every tool listed. Bid on the milestones where you have hands-on experience.

Alternative tools (Loki, Tempo, Zipkin, Datadog, etc.) are welcome — I care about learning the concepts, not being locked to a specific product.

Fixed price per milestone. Each delivers a working, containerised POC I can run locally via docker-compose.

MILESTONE 1 — Structured Logging (Seq or ELK Stack)

.NET 8 Web API with Serilog → Seq or ELK. Enrichers (correlation ID, request context), proper log levels, a dashboard showing search/filter/correlation, docker-compose with all services, brief walkthrough.

MILESTONE 2 — Distributed Tracing (Jaeger or Zipkin)

OpenTelemetry SDK configured for HTTP, DB, and custom spans. Multi-service demo showing end-to-end trace propagation. Jaeger/Zipkin in Docker with traces flowing. Examples of finding slow ops and failures.

MILESTONE 3 — Metrics & Alerting (Prometheus + Grafana)

Prometheus scraping .NET metrics (request rate, error rate, P95 latency, GC/thread pool). Grafana dashboard with meaningful panels. Alert rules for error rate spikes, latency thresholds, health failures. AlertManager/Grafana notification channel configured.

MILESTONE 4 — Health Checks & Proactive Monitoring

ASP.NET Health Checks for dependencies (DB, external APIs, message queues, disk/memory). Health check UI. Prometheus/Grafana integration for tracking status over time. Demo of degraded dependency triggering an alert before users notice.

MILESTONE 5 — Containerised Deployment & Full Integration

Single docker-compose running everything:

.NET API, logging stack, Jaeger, Prometheus, Grafana with pre-provisioned dashboards.

Demo scenario:

trigger a failure → watch it flow through logs → traces → metrics → alert. Architecture README + Kubernetes migration guidance.

Contract duration of 1 to 3 months. with None hours per week.

Mandatory skills:

Loki, Zipkin, Datadog, ELK Stack, OpenTelemetry , Jaeger, Prometheus , Grafana, Elasticsearch