diff --git a/.jules/bolt.md b/.jules/bolt.md
index 289fae6..3ac930d 100644
--- a/.jules/bolt.md
+++ b/.jules/bolt.md
@@ -23,3 +23,7 @@
 ## 2026-09-19 - Avoid chained array methods before Set initialization
 **Learning:** Chaining `.map().filter()` inside a `new Set()` constructor creates multiple intermediate arrays, causing unnecessary memory allocation and garbage collection overhead in hot paths (like ticker normalization during discovery).
 **Action:** Replace chained array methods with a direct indexed `for` loop that transforms, validates, and `.add()`s directly to a freshly initialized `Set`.
+
+## 2026-09-30 - Avoid chained array operations during unique subset generation
+**Learning:** Generating a subset of unique elements by mapping arrays, concatenating via spread operators into a Set, converting back via Array.from(), and slicing creates multiple intermediate objects and unneeded iterations. This creates excessive garbage collection, particularly when dealing with large inner loops like backtesting tickers.
+**Action:** When extracting a unique, bounded subset from multiple arrays, use an explicitly bounded `for` loop and track seen elements with a `Set` to terminate iteration exactly when the limit is reached, skipping allocations for discarded elements.
diff --git a/src/agent/backtestRunner.ts b/src/agent/backtestRunner.ts
index e53b03c..9a63a7c 100644
--- a/src/agent/backtestRunner.ts
+++ b/src/agent/backtestRunner.ts
@@ -347,10 +347,25 @@ export async function runBacktestSession(
           universe: BACKTEST_DISCOVERY_UNIVERSE,
           limit: maxCandidates,
         });
-        const held = (await broker.snapshot()).positions.map((p) => p.ticker);
-        const tickers = Array.from(
-          new Set([...held, ...discovery.candidates.map((c) => c.ticker)]),
-        ).slice(0, maxCandidates);
+        const snapshotPositions = (await broker.snapshot()).positions;
+        const tickers: string[] = [];
+        const seen = new Set<string>();
+
+        // ⚡ Bolt: Optimize ticker extraction by avoiding chained .map(), Set spread, and .slice() allocations
+        for (let i = 0; i < snapshotPositions.length && tickers.length < maxCandidates; i++) {
+          const ticker = snapshotPositions[i]?.ticker;
+          if (ticker && !seen.has(ticker)) {
+            seen.add(ticker);
+            tickers.push(ticker);
+          }
+        }
+        for (let i = 0; i < discovery.candidates.length && tickers.length < maxCandidates; i++) {
+          const ticker = discovery.candidates[i]?.ticker;
+          if (ticker && !seen.has(ticker)) {
+            seen.add(ticker);
+            tickers.push(ticker);
+          }
+        }
 
         const decisions: FinalDecision[] = [];
         for (const ticker of tickers) {