OpenTelemetry Certified Associate
Most courses teach you to
pass the exam.
This course teaches you to understand OTel.
The OTCA certification is a natural next step.
What you'll be able to do after this course
- Explain why OTel exists and what problem it solves (interview-ready)
- Instrument a Python, Node.js, and Java service with OTel
- Write a complete OTel Collector config with multiple pipelines
- Debug a broken trace across three services from symptoms alone
- Design a sampling strategy with real cost/accuracy tradeoffs
- Pass the OTCA exam with confidence
What you'll learn
- 0Start →Foundation Free 30 min
Foundation: Why Telemetry Fails Without Standards
Before OTel existed, every vendor had its own SDK, its own protocol, its own data model. Understand what problem OTel actually solves — and why the architecture is the way it is.
- 0Start →Foundation Free 25 min
Observability 101: Signals, Layers, and Why Context Wins
What traces, metrics, and logs actually are, what questions each signal answers, how telemetry flows from code to backend, and why the three signals are worth more together than apart.
- 0Start →Reference Free 20 min
InsureWatch — Application Guide
Architecture, service map, and instrumentation overview for the InsureWatch lab application. Read this before starting any lab.
- 1Start →Module Free 60 min
Module 1: Signals — The Unified Data Model
Traces, metrics, and logs look like three different things. They share a common context schema that makes them one correlated stream — if you set it up right. Here is what each signal actually is.
- 2Module 60 min
Module 2: Context Propagation
Context propagation is automatic — until it isn't. Understand the W3C TraceContext standard, where propagation breaks across service boundaries, and how to fix it.
- 2Lab 45 min
Lab 1: Find the Broken Trace
InsureWatch is running but traces are fragmented. Context propagation is misconfigured in claims-service. Find the mismatch, understand why it breaks the trace, and fix it.
- 3Module 90 min
Module 3: Instrumentation — SDK, API, Auto vs Manual
Auto-instrumentation covers known libraries. Your business logic is invisible to it. Learn the API/SDK split, provider initialization, and how to manually instrument what the agent can't see.
- 3Lab 60 min
Lab 2: Instrument the Black Box
claims-service is healthy but completely invisible. No spans, no log correlation, no database visibility. Re-add auto-instrumentation, then write manual spans with meaningful attributes.
- 4Module 45 min
Module 4: Semantic Conventions
Semantic conventions are not optional naming suggestions. Break them and every downstream tool that expects standard attribute names breaks silently. They are a portability contract.
- 5Module 90 min
Module 5: The Collector — Architecture and Pipelines
Without a Collector, the backend URL lives in every service config. You cannot sample, transform, or switch backends without redeploying everything. Here is how the Collector actually works.
- 5Lab 45 min
Lab 3: Wire the Collector Pipeline
An OTel Collector sits between InsureWatch and Grafana, but its pipeline is empty. Complete the config — receivers, processors, exporters — and get telemetry flowing again.
- 6Lab 90 min
Lab 4: Restore Full Observability
Three instrumentation problems are active simultaneously: a broken Collector pipeline, a propagator mismatch, and missing instrumentation. Diagnose and fix all three to restore complete end-to-end visibility.
The Chaos Lab — industry-unique
Lab 4 uses a live chaos controller on InsureWatch (Python/Node.js/Java) to inject one of three production failure scenarios. You don't know which. Use traces, metrics, and logs to find it — no hints, no step-by-step instructions. Exactly like a real incident. No other OTCA prep course has this.
Zero-friction lab environment
All labs run locally. No cloud account required for Labs 1–3.
docker run -p 3000:3000 -p 4317:4317 -p 4318:4318 grafana/otel-lgtm
One command. Grafana UI at localhost:3000. Traces, metrics, and logs — all correlated.
Simple, one-time pricing
No subscription. Lifetime access. PPP discounts applied automatically.
Core Course
$49 USD
- All 8 modules + Foundation
- Labs 1–4 (incl. Chaos Lab)
- InsureWatch lab app
- Lifetime access + updates
Modules 1–5 + Labs 1–3 available now. Modules 6–8 + Chaos Lab ship within 6 weeks, added automatically.
Buy Core CourseExam Bundle
$89 USD
- Everything in Core
- 30 OTCA practice questions
- Domain study checklist
- 48-hour rapid review guide
Modules 1–5 + Labs 1–3 available now. Modules 6–8 + Chaos Lab ship within 6 weeks, added automatically.
Buy Exam Bundle
otel.guru is not affiliated with Grafana Labs, Uptrace, or any backend vendor in these labs.
PPP discounts applied automatically at checkout. Prices shown in USD.
By purchasing, you agree to our Terms & Conditions. Course access is personal and non-transferable. Certification outcomes are not guaranteed.
FAQ
- Is this only for people taking the OTCA exam?
- No. The course is designed to build genuine OTel understanding. The OTCA certification is a natural next step for those who want to validate their skills. If you want to instrument services, operate observability pipelines, or debug OTel issues in production — this course applies directly.
- Do I need Kubernetes experience?
- No. Kubernetes is covered conceptually in Module 6 (because the exam tests it). The hands-on labs run on Docker Compose — 8GB RAM and Docker Desktop is all you need.
- What is InsureWatch?
- InsureWatch is a purpose-built multi-language microservices app (Python, Node.js, Java) used as the lab environment. It includes a chaos controller that injects real failure scenarios. You clone it from GitHub — it's open source.
- What if I'm a complete beginner to observability?
- The course assumes you know what traces, metrics, and logs are at a surface level — you've built dashboards, worked with tools like Prometheus or Datadog. If you've never touched observability, we'd suggest a quick primer first.
- When will Modules 6–8 be available?
- Within 6 weeks of this launch. All purchasers receive them automatically — no re-purchase, no extra steps.