lading-optimize-hunt

Optimization Hunt

Coordinates optimization attempts: captures baselines, implements changes, invokes review, and records all outcomes.

Role: Coordinator and Recorder

Hunt is the coordinator and recorder — it captures baselines, implements changes, hands off to review, and records all outcomes.

Hunt does NOT:

• Run post-change benchmarks (review does this)
• Make pass/fail decisions on optimizations (review does this)

Hunt DOES:

• Record all verdicts and outcomes in .claude/skills/lading-optimize-hunt/assets/db.yaml after review returns

---

Phase 0: Pre-flight

Run /lading-preflight.

---

Phase 1: Find Target

Run /lading-optimize-find-target.

It returns a YAML block with 6 fields: pattern, technique, target, file, bench, fingerprint - Print it out.

---

Phase 2: Establish Baseline

CRITICAL: Capture baseline metrics BEFORE making any code changes.

Identify the Benchmark Target

Use the bench and fingerprint fields from find-target's output — they are repo-relative paths ready to use:

BENCH=<bench field without extension>   # e.g. from "lading_payload/benches/syslog.rs" use "--bench syslog"
PAYLOADTOOL_CONFIG=<fingerprint field>  # e.g. "ci/fingerprints/syslog/lading.yaml"

Stage 1: Clear previous benchmarks

Clear any previously captured baselines so stale data cannot contaminate this run.

rm -f /tmp/criterion-baseline.log /tmp/baseline.json /tmp/baseline-mem.txt
rm -rf target/criterion

Stage 2: Micro-benchmark Baseline

Run only the benchmark for your target:

cargo criterion --bench "$BENCH" 2>&1 | tee /tmp/criterion-baseline.log

Stage 3: Macro-benchmark Baseline

Use the matching fingerprint config:

cargo build --release --bin payloadtool
hyperfine --warmup 3 --runs 30 --export-json /tmp/baseline.json \
  "./target/release/payloadtool $PAYLOADTOOL_CONFIG"

./target/release/payloadtool "$PAYLOADTOOL_CONFIG" --memory-stats 2>&1 | tee /tmp/baseline-mem.txt