playbooks

home / skills / sandraschi / advanced-memory-mcp / performance-tuning-expert

performance-tuning-expert skill

/skills/technical/performance-tuning-expert

This skill helps optimize application performance by guiding profiling, backend/frontend tuning, and scalable infrastructure strategies with actionable

npx playbooks add skill sandraschi/advanced-memory-mcp --skill performance-tuning-expert

Review the files below or copy the command above to add this skill to your agents.

Files (11)
SKILL.md
2.3 KB
---
name: performance-tuning-expert
description: Application performance specialist for profiling, optimization, and scaling strategies
license: Proprietary
---

# Performance Tuning Expert
> **Status**: ✅ Research complete
> **Last validated**: 2025-11-08
> **Confidence**: 🟡 Medium — Research-backed tuning playbook – review semi-annually

## How to use this skill
1. Triage performance goals via [modules/core-guidance.md](modules/core-guidance.md).
2. Instrument and profile using [modules/profiling-and-metrics.md](modules/profiling-and-metrics.md).
3. Apply backend optimizations from [modules/backend-optimization.md](modules/backend-optimization.md).
4. Improve user-facing performance using [modules/frontend-and-client.md](modules/frontend-and-client.md).
5. Tune runtime and infrastructure with [modules/infrastructure-and-runtime.md](modules/infrastructure-and-runtime.md).
6. Lock improvements through [modules/regression-testing-and-guardrails.md](modules/regression-testing-and-guardrails.md).
7. Track additional research in [modules/known-gaps.md](modules/known-gaps.md) and follow the cadence in [modules/research-checklist.md](modules/research-checklist.md).

## Module overview
- [Core guidance](modules/core-guidance.md) — intake checklist, prioritization criteria, ROI framing.
- [Profiling & metrics](modules/profiling-and-metrics.md) — observability setup, profiling stacks, benchmarking.
- [Backend optimization](modules/backend-optimization.md) — language/runtime tuning, data access improvements.
- [Frontend & client](modules/frontend-and-client.md) — Core Web Vitals, mobile optimization, offline strategies.
- [Infrastructure & runtime](modules/infrastructure-and-runtime.md) — caching, autoscaling, hardware selection.
- [Regression testing & guardrails](modules/regression-testing-and-guardrails.md) — load testing, CI integration, alerting.
- [Known gaps](modules/known-gaps.md) — ongoing research backlog.
- [Research checklist](modules/research-checklist.md) — semi-annual review plan.

## Research status
- Material incorporates latest Web Vitals, Netflix/AWS performance engineering guidance, and mechanical sympathy insights.
- Next validation due 2026-05-01 or earlier if significant runtime updates land (e.g., JVM, Node, Rust).
- Known gaps cover ML/AI workloads and eBPF observability patterns pending deeper study.

Overview

This skill is an application performance specialist that guides profiling, optimization, and scaling strategies across backend, frontend, and infrastructure. It codifies a practical playbook for triage, instrumentation, targeted fixes, and regression protection. The guidance is research-backed and designed for repeatable improvements across languages and runtimes.

How this skill works

Start by defining clear performance goals and an ROI-driven prioritization to focus effort where it matters most. Instrument the system with observability and profiling tools, run targeted benchmarks to identify hotspots, and apply language- and platform-specific optimizations. Finish by hardening gains with load testing, CI guardrails, and continuous measurement to prevent regressions.

When to use it

  • When user-facing latency or Core Web Vitals are failing to meet targets
  • When backend services show CPU, memory, or I/O hotspots under production load
  • Before major releases or infrastructure changes to prevent regressions
  • When planning capacity, autoscaling, or hardware upgrades
  • During postmortems for performance incidents to find root causes

Best practices

  • Always start with measurable goals (SLOs/SLIs) and quantify expected ROI before tuning
  • Prefer observability-first: collect traces, metrics, and profiles before changing code
  • Use targeted microbenchmarks and controlled load tests to validate improvements
  • Treat performance work as testable features: add regression tests and CI checks
  • Balance optimization with maintainability; focus on high-impact, low-risk changes

Example use cases

  • Reducing 95th-percentile API latency by profiling request handlers and optimizing database access patterns
  • Improving Core Web Vitals by deferring noncritical JavaScript and optimizing resource loading
  • Scaling a service by introducing caching tiers, connection pooling, and tuned autoscaling policies
  • Diagnosing memory leaks in long-running processes via heap profiling and GC tuning
  • Preventing regressions with performance budgets integrated into CI and synthetic monitoring

FAQ

How do I pick the first area to optimize?

Start with a business-impact analysis: pick the flow that affects most users or costs most in infrastructure, then gather metrics and profile that flow first.

Which tools should I use for profiling?

Use language-appropriate profilers and distributed tracing for system-wide visibility; complement with APM, flamegraphs, and benchmark suites for targeted hotspots.