Programme objective: You plan a performance study, use artificial intelligence as a drafting and detection aid, run the study on a system you are allowed to test, read the result against a target you wrote first, and, in an advanced module, rehearse a controlled failure inside an isolated authorised environment. You own the run, the interpretation, and the release decision.
Learner path: Fundamentals, strategy and AI-assisted workload modelling, Apache JMeter with AI-assisted script drafting from journeys, advanced JMeter, LoadRunner where you have the licence, scenarios, API performance, database performance, AI analysis, observability, optimisation, reporting, then chaos engineering with AI-assisted resilience analysis. This is one programme, not a separate chaos course. The surrounding gates are on Complete AI-Native SDLC Engineering.
Suggested time inside the pathway is 25 hours, within a 16-week professional programme of about 160 to 200 hours. That shape is subject to batch and faculty planning. The cohort and any certificate are confirmed when you enquire. This page does not state fees, batch dates, faculty names, or placement rates. A classroom run is not a capacity promise.
Module 1 — Performance fundamentals
You name the question the run must answer, the activity under study, and the ceiling you will not pass.
Module 2 — Performance strategy and workload modelling
You write who uses the system, what finished means, and which reading will decide the outcome, before any chart exists. An assistant may draft a workload model from the journeys you listed. You accept, rewrite, or reject that draft before the run.
Module 3 — Apache JMeter
You build the open-source laboratory in Apache JMeter and run it only against the system the exercise names. An assistant may draft a script skeleton from a journey. You review every sampler, assertion, and pacing value before you execute.
Module 4 — Advanced JMeter
You extend that JMeter laboratory with the data, pacing, and checks the strategy already listed, and you stop at the agreed ceiling.
Module 5 — LoadRunner
You study LoadRunner only where you already have the licensed environment. JMeter remains the open-source laboratory for the cohort.
Module 6 — Test scenarios
You design load, stress, spike, endurance, scalability, and volume scenarios, each with a ceiling written before the run. AI may propose mix and ramp shapes. You keep the ceiling and the stop condition you wrote.
Module 7 — API performance
You measure the application programming interface the strategy named, and you say which number answers the question you wrote down. An assistant may draft a slow-API investigation note. You tie every cause to evidence from that run.
Module 8 — Database performance
You look at the database during the same run, including the connection pool, and you tie a slow step to evidence from that run. An assistant may suggest a slow-query line of enquiry. You confirm it against the same metrics and traces.
Module 9 — AI performance analysis
You use the assistant for anomaly detection, for correlating logs, traces, and metrics, and for draft comparison reports between baseline and candidate builds. You correct any figure or cause the table does not support. Capacity recommendations stay drafts until a person validates them.
Module 10 — Observability
You read a run beside Prometheus, Grafana, and OpenTelemetry signals. AI may suggest which log, trace, and metric belong together for a slow step. You say which signal belongs to the activity you tested.
Module 11 — Performance optimization
You propose a change that the evidence supports, then you rerun and compare the new reading with the target. An assistant may draft the before-and-after comparison. You do not treat one chart as a finished tuning claim.
Module 12 — Performance reporting
You write what the run showed, what it does not show, and the comparison with the agreed target. AI may draft the report and a capacity recommendation. You own the interpretation and the release recommendation that goes to the person who owns the system.
Module 13 — Chaos engineering
This advanced module stays on this page. You study fundamentals, infrastructure faults, network faults, application faults, database resilience, microservice resilience, Kubernetes chaos, recovery validation, and AI-assisted analysis of chaos or resilience results. The tools you name are Chaos Mesh, LitmusChaos, and Toxiproxy. Every experiment is in an isolated authorised environment with a predefined blast radius and a rollback. An assistant may summarise recovery signals. You validate that the service recovered and that the blast radius stayed inside the plan. This page does not give fault-injection commands.
Expected results for ordinary cases are AI-Powered Test Automation Engineering. The application you build is AI-Native Software Engineering and Development. Authorised defensive testing is AI-Powered Application, API and LLM Security Testing.
