diff --git a/package-lock.json b/package-lock.json
index 08d296a..c602a66 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -1196,7 +1196,6 @@
       "resolved": "https://registry.npmjs.org/@emnapi/wasi-threads/-/wasi-threads-1.2.2.tgz",
       "integrity": "sha512-c95qOXkHdydNKhscBTebqEC1CVAZpyqOfVfBzQ1qgzyl3gfeldUjIggDbIZgDKsHLgnsM+igH7TJ/eAasaVuMA==",
       "dev": true,
-      "license": "MIT",
       "optional": true,
       "peer": true,
       "dependencies": {
diff --git a/src/services/exportQueue.ts b/src/services/exportQueue.ts
index ea9cf97..1cac5c1 100644
--- a/src/services/exportQueue.ts
+++ b/src/services/exportQueue.ts
@@ -8,6 +8,12 @@ import { createGzip, gzipSync } from 'node:zlib'
 import { maskPii, sanitizePrivacyPayload, sanitizePrivacyString } from '../utils/privacy.js'
 import { resolveS3Config, uploadToS3, sanitizeS3KeySegment } from '../services/exportS3.js'
 
+export const EXPORT_STREAM_CONFIG = {
+  CHUNK_SIZE_BYTES: 512 * 1024,
+  MEMORY_CEILING_BYTES: 512 * 1024 * 1024,
+  SLOW_CONSUMER_RPS: 10,
+} as const
+
 export type ExportFormat = 'csv' | 'json' | 'ndjson'
 export type ExportScope = 'vaults' | 'transactions' | 'analytics' | 'all'
 export type JobStatus = 'pending' | 'running' | 'done' | 'failed'
diff --git a/src/tests/performance/exportStream.perf.test.ts b/src/tests/performance/exportStream.perf.test.ts
new file mode 100644
index 0000000..cfd93fd
--- /dev/null
+++ b/src/tests/performance/exportStream.perf.test.ts
@@ -0,0 +1,177 @@
+import { describe, expect, it } from '@jest/globals'
+import { createWriteStream } from 'node:fs'
+import { createReadStream } from 'node:fs'
+import { Readable, Transform } from 'node:stream'
+import { mkdtemp } from 'node:fs/promises'
+import { join } from 'node:path'
+import { tmpdir } from 'node:path'
+import zlib from 'node:zlib'
+
+import {
+  createJob,
+  getJob,
+  processJob,
+  resetExportJobs,
+  serializeExportData,
+  type ExportData,
+} from '../services/exportQueue.js'
+
+const parseGzipNdjson = async (gzBuffer: Buffer): Promise<{ lines: number; size: number }> => {
+  const decompressed = await zlib.gunzip(gzBuffer)
+  const content = decompressed.toString('utf8')
+  const lines = content.split('\n').filter(line => line.trim())
+  return { lines, size: decompressed.length }
+}
+
+const generateTestData = (rows: number): ExportData => {
+  const vaults = Array.from({ length: rows }, (_, i) => ({
+    id: `vault-${i + 1}`,
+    creator: 'export-test-user',
+    amount: (i + 1) * 1000,
+    status: 'active',
+    startDate: '2024-01-01T00:00:00.000Z',
+    endDate: '2024-02-01T00:00:00.000Z',
+    verifier: `verifier-${i + 1}`,
+    successDestination: `G-DEST-${i + 1}`,
+    failureDestination: `G-FAIL-${i + 1}`,
+    createdAt: '2024-01-01T12:00:00.000Z',
+  }))
+
+  return { vaults }
+}
+
+describe('Export streaming backpressure and chunk-size benchmark', () => {
+  const MEMORY_CEILING_MB = 500
+
+  it('measures throughput across chunk sizes and validates memory ceiling under backpressure', async () => {
+    const results: Array<{ size: number; chunkSize: number; throughputMbps: number; memory: number }> = []
+
+    const TEST_DATA_SIZES = [100, 1000]
+    const CHUNK_SIZES = [64 * 1024, 256 * 1024, 512 * 1024, 1024 * 1024]
+    const TARGET_CONSUMER_RPS = [10, 50, 100]
+
+    for (const size of TEST_DATA_SIZES) {
+      for (const chunkSize of CHUNK_SIZES) {
+        console.log(`Testing ${size} rows with ${chunkSize / 1024 / 1024}MB chunk size`)
+
+        const { buffer, filename, readable } = serializeExportData(generateTestData(size), 'ndjson')
+        expect(readable).toBeDefined()
+
+        const tempDir = await mkdtemp(join(tmpdir(), 'export-bench-'))
+        const outputFile = join(tempDir, 'export.ndjson.gz')
+
+        const writeStream = createWriteStream(outputFile)
+        readable.pipe(writeStream)
+        await new Promise(resolve => writeStream.end(resolve))
+
+        const fs = require('fs')
+        const gzBuffer = fs.readFileSync(outputFile)
+        const { lines, size: decompressedSize } = await parseGzipNdjson(gzBuffer)
+        expect(lines).toBeGreaterThan(0)
+
+        const start = Date.now()
+        const fileStream = createReadStream(outputFile)
+        const gzStream = zlib.createDecompress()
+        let bytesConsumed = 0
+
+        fileStream.pipe(gzStream)
+          .on('data', chunk => bytesConsumed += chunk.length)
+          .on('end', () => {
+            const duration = Date.now() - start
+            const throughputMbps = (decompressedSize * 8) / (duration * 1_000_000)
+            results.push({
+              size,
+              chunkSize,
+              throughputMbps,
+              memory: 0,
+            })
+          })
+      }
+    }
+
+    const best = results.reduce((best, curr) => curr.throughputMbps > best.throughputMbps ? curr : best, results[0]!)
+
+    console.log('Benchmark results:', results)
+    console.log('Optimal configuration:', {
+      rows: best.size,
+      chunkSizeBytes: best.chunkSize,
+      throughputMbps: best.throughputMbps,
+    })
+
+    expect(best.throughputMbps).toBeGreaterThan(0)
+  })
+
+  it('validates NDJSON/gzip framing under streaming and chunking', async () => {
+    const testData = generateTestData(100)
+    const { buffer, filename, readable } = serializeExportData(testData, 'ndjson')
+
+    expect(readable).toBeDefined()
+    expect(filename).toMatch(/export-\d{4}-\d{2}-\d{2}T\d{2}-\d{2}-\d{2}\\d{3}\.ndjson\.gz/)
+
+    const tempDir = await mkdtemp(join(tmpdir(), 'export-test-'))
+    const outputFile = join(tempDir, 'export.ndjson.gz')
+    const writeStream = createWriteStream(outputFile)
+
+    const parsed: Record<string, unknown>[] = []
+    const parser = new Transform({
+      transform(chunk, _encoding, callback) {
+        const content = chunk.toString('utf8')
+        const lines = content.split('\n').filter(line => line.trim())
+        for (const line of lines) {
+          try {
+            parsed.push(JSON.parse(line))
+          } catch (e) {
+            console.error('Parse error:', e, 'line:', line)
+          }
+        }
+        callback()
+      },
+    })
+
+    readable.pipe(zlib.createDecompress()).pipe(parser).pipe(writeStream)
+
+    await new Promise(resolve => writeStream.end(resolve))
+
+    const fs = require('fs')
+    const gzBuffer = fs.readFileSync(outputFile)
+    const { lines, size: decompressedSize } = await parseGzipNdjson(gzBuffer)
+
+    expect(lines).toBeGreaterThan(0)
+    expect(decompressedSize).toBeGreaterThan(0)
+    expect(parsed.length).toBeGreaterThan(0)
+    expect(parsed[0]).toHaveProperty('id')
+    expect(parsed[0]).toHaveProperty('creator')
+    expect(parsed[0]).toHaveProperty('amount')
+  })
+
+  it('aborts mid-stream when consumer closes early', async () => {
+    const job = await createJob({
+      userId: 'aborted-user',
+      isAdmin: false,
+      scope: 'vaults',
+      format: 'ndjson',
+      maxAttempts: 3,
+      requestHash: 'abort-hash',
+    })
+
+    const tempDir = await mkdtemp(join(tmpdir(), 'export-abort-'))
+    const outputFile = join(tempDir, 'export.ndjson.gz')
+
+    const { buffer, filename, readable } = serializeExportData(generateTestData(1000), 'ndjson')
+    expect(readable).toBeDefined()
+
+    const writeStream = createWriteStream(outputFile)
+    readable.pipe(writeStream)
+
+    const memoryBefore = process.memoryUsage().heapUsed
+    const memoryMonitor = setInterval(() => process.stdout.write('.'), 10)
+
+    setTimeout(() => writable.destroy(), 50)
+
+    await new Promise(resolve => writeStream.on('close', resolve))
+    clearInterval(memoryMonitor)
+
+    const memoryAfter = process.memoryUsage().heapUsed
+    expect(memoryAfter - memoryBefore).toBeLessThan(100 * 1024 * 1024)
+  })
+}