diff --git a/servers/legion/apps/neuron-gamma.yaml b/servers/legion/apps/neuron-gamma.yaml
new file mode 100644
index 0000000..7524e60
--- /dev/null
+++ b/servers/legion/apps/neuron-gamma.yaml
@@ -0,0 +1,89 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: neuron-gamma
+  namespace: neuron
+  labels:
+    app: neuron-gamma
+spec:
+  replicas: 1
+  strategy:
+    type: Recreate
+  selector:
+    matchLabels:
+      app: neuron-gamma
+  template:
+    metadata:
+      labels:
+        app: neuron-gamma
+    spec:
+      containers:
+        - name: neuron
+          image: registry.neuralplatform.ai/neural-platform/neuron:latest
+          imagePullPolicy: Always
+          command:
+            - bash
+            - -c
+            - mkdir -p /root/.neuron && ln -sf /data/neuron.db /root/.neuron/neuron.db && python -m synapse platform serve --mcp-host 0.0.0.0 --mcp-port 8001 --webhook-port 3847
+          ports:
+            - name: mcp
+              containerPort: 8001
+            - name: webhook
+              containerPort: 3847
+          envFrom:
+            - configMapRef:
+                name: neuron-config
+            - secretRef:
+                name: neuron-secrets
+          volumeMounts:
+            - name: data
+              mountPath: /data
+            - name: app-config
+              mountPath: /app/config/default.yaml
+              subPath: default.yaml
+          resources:
+            requests:
+              memory: 256Mi
+              cpu: 100m
+              ephemeral-storage: 256Mi
+            limits:
+              memory: 1Gi
+              cpu: 500m
+              ephemeral-storage: 1Gi
+          livenessProbe:
+            tcpSocket:
+              port: mcp
+            initialDelaySeconds: 15
+            periodSeconds: 30
+          readinessProbe:
+            tcpSocket:
+              port: mcp
+            initialDelaySeconds: 10
+            periodSeconds: 10
+      volumes:
+        - name: data
+          persistentVolumeClaim:
+            claimName: neuron-gamma-data
+        - name: app-config
+          configMap:
+            name: neuron-config
+            items:
+              - key: default.yaml
+                path: default.yaml
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: neuron-gamma
+  namespace: neuron
+spec:
+  selector:
+    app: neuron-gamma
+  ports:
+    - name: mcp
+      port: 8001
+      targetPort: 8001
+    - name: webhook
+      port: 3847
+      targetPort: 3847
+  type: ClusterIP
diff --git a/servers/legion/k8s/neuron/ingress.yaml b/servers/legion/k8s/neuron/ingress.yaml
index aac1b56..7c416d1 100644
--- a/servers/legion/k8s/neuron/ingress.yaml
+++ b/servers/legion/k8s/neuron/ingress.yaml
@@ -12,6 +12,15 @@ spec:
     - secretName: neuron-tls
       hosts:
         - neuron.neuralplatform.ai
+    - secretName: neuron-alpha-tls
+      hosts:
+        - neuron-alpha.neuralplatform.ai
+    - secretName: neuron-beta-tls
+      hosts:
+        - neuron-beta.neuralplatform.ai
+    - secretName: neuron-gamma-tls
+      hosts:
+        - neuron-gamma.neuralplatform.ai
   rules:
     - host: neuron.neuralplatform.ai
       http:
@@ -23,3 +32,33 @@ spec:
                 name: neuron
                 port:
                   number: 8001
+    - host: neuron-alpha.neuralplatform.ai
+      http:
+        paths:
+          - path: /
+            pathType: Prefix
+            backend:
+              service:
+                name: neuron-alpha
+                port:
+                  number: 8001
+    - host: neuron-beta.neuralplatform.ai
+      http:
+        paths:
+          - path: /
+            pathType: Prefix
+            backend:
+              service:
+                name: neuron-beta
+                port:
+                  number: 8001
+    - host: neuron-gamma.neuralplatform.ai
+      http:
+        paths:
+          - path: /
+            pathType: Prefix
+            backend:
+              service:
+                name: neuron-gamma
+                port:
+                  number: 8001
diff --git a/servers/legion/k8s/neuron/pvc-gamma.yaml b/servers/legion/k8s/neuron/pvc-gamma.yaml
new file mode 100644
index 0000000..1533c02
--- /dev/null
+++ b/servers/legion/k8s/neuron/pvc-gamma.yaml
@@ -0,0 +1,12 @@
+apiVersion: v1
+kind: PersistentVolumeClaim
+metadata:
+  name: neuron-gamma-data
+  namespace: neuron
+spec:
+  storageClassName: local-path
+  accessModes:
+    - ReadWriteOnce
+  resources:
+    requests:
+      storage: 5Gi
diff --git a/servers/legion/k8s/neuron/snapshot-job.yaml b/servers/legion/k8s/neuron/snapshot-job.yaml
new file mode 100644
index 0000000..3c400a6
--- /dev/null
+++ b/servers/legion/k8s/neuron/snapshot-job.yaml
@@ -0,0 +1,44 @@
+# Snapshot Job — copies prod DB into an experiment slot via sqlite3 .backup
+# Usage: set SLOT to alpha, beta, or gamma before applying
+#
+#   SLOT=alpha && sed "s/SLOT/${SLOT}/g" snapshot-job.yaml | kubectl apply -f -
+#
+# The job completes and self-deletes after 5 minutes.
+# Always snapshot before deploying a new experiment to a slot.
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: snapshot-prod-to-SLOT
+  namespace: neuron
+spec:
+  ttlSecondsAfterFinished: 300
+  template:
+    spec:
+      restartPolicy: Never
+      initContainers:
+        - name: stop-check
+          image: registry.neuralplatform.ai/neural-platform/neuron:latest
+          command: ["/bin/sh", "-c", "echo 'Snapshotting prod DB to SLOT — ensure neuron-SLOT pod is stopped first'"]
+      containers:
+        - name: snapshot
+          image: keinos/sqlite3:latest
+          command: ["/bin/sh", "-c"]
+          args:
+            - |
+              echo "Starting backup from prod to SLOT..."
+              sqlite3 /src/neuron.db ".backup /dst/neuron.db"
+              echo "Done. Prod DB snapshotted to SLOT."
+              ls -lh /dst/neuron.db
+          volumeMounts:
+            - name: src
+              mountPath: /src
+              readOnly: true
+            - name: dst
+              mountPath: /dst
+      volumes:
+        - name: src
+          persistentVolumeClaim:
+            claimName: neuron-data
+        - name: dst
+          persistentVolumeClaim:
+            claimName: neuron-SLOT-data
diff --git a/servers/legion/main.tf b/servers/legion/main.tf
index cd83305..e02a6a7 100644
--- a/servers/legion/main.tf
+++ b/servers/legion/main.tf
@@ -52,7 +52,7 @@ resource "cloudflare_zero_trust_tunnel_cloudflared_config" "legion" {
   tunnel_id  = var.cloudflare_tunnel_id
 
   config {
-    # neuron.neuralplatform.ai — MCP server (SSE transport, requires CF-Access headers)
+    # neuron.neuralplatform.ai — MCP server, stable prod
     ingress_rule {
       hostname = "neuron.neuralplatform.ai"
       service  = "https://traefik.kube-system.svc:443"
@@ -61,6 +61,33 @@ resource "cloudflare_zero_trust_tunnel_cloudflared_config" "legion" {
       }
     }
 
+    # neuron-alpha.neuralplatform.ai — experiment zone 1 (CI auto-deploys here)
+    ingress_rule {
+      hostname = "neuron-alpha.neuralplatform.ai"
+      service  = "https://traefik.kube-system.svc:443"
+      origin_request {
+        no_tls_verify = true
+      }
+    }
+
+    # neuron-beta.neuralplatform.ai — experiment zone 2
+    ingress_rule {
+      hostname = "neuron-beta.neuralplatform.ai"
+      service  = "https://traefik.kube-system.svc:443"
+      origin_request {
+        no_tls_verify = true
+      }
+    }
+
+    # neuron-gamma.neuralplatform.ai — experiment zone 3
+    ingress_rule {
+      hostname = "neuron-gamma.neuralplatform.ai"
+      service  = "https://traefik.kube-system.svc:443"
+      origin_request {
+        no_tls_verify = true
+      }
+    }
+
     ingress_rule {
       hostname = "vault.neuralplatform.ai"
       service  = "https://traefik.kube-system.svc:443"
diff --git a/servers/legion/scripts/neuron-self-improve.sh b/servers/legion/scripts/neuron-self-improve.sh
new file mode 100755
index 0000000..1a9d16d
--- /dev/null
+++ b/servers/legion/scripts/neuron-self-improve.sh
@@ -0,0 +1,235 @@
+#!/usr/bin/env bash
+# neuron-self-improve.sh — self-improvement loop for Neuron
+#
+# Manages code experiments across alpha/beta/gamma zones and promotes winners to prod.
+# All cluster changes go through git → Argo CD. No direct kubectl mutations.
+#
+# Usage:
+#   ./neuron-self-improve.sh snapshot <slot>          # copy prod DB to slot
+#   ./neuron-self-improve.sh deploy <slot> <sha>      # point slot at a specific image
+#   ./neuron-self-improve.sh promote <slot>           # copy slot's image to prod
+#   ./neuron-self-improve.sh status                   # show what's running in each zone
+#   ./neuron-self-improve.sh rollback <slot> <sha>    # revert slot to a previous image
+
+set -euo pipefail
+
+INFRA_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+REGISTRY="registry.neuralplatform.ai/neural-platform/neuron"
+KUBECONFIG="${KUBECONFIG:-$HOME/.kube/legion-config}"
+
+# ── helpers ───────────────────────────────────────────────────────────────────
+
+slot_manifest() {
+  local slot=$1
+  if [[ "$slot" == "prod" ]]; then
+    echo "$INFRA_DIR/servers/legion/apps/neuron.yaml"
+  else
+    echo "$INFRA_DIR/servers/legion/apps/neuron-${slot}.yaml"
+  fi
+}
+
+current_sha() {
+  local slot=$1
+  grep "image:" "$(slot_manifest "$slot")" | grep -oP '[a-f0-9]{40}' | head -1
+}
+
+set_image() {
+  local slot=$1 sha=$2
+  local manifest
+  manifest="$(slot_manifest "$slot")"
+  sed -i "s|${REGISTRY}:.*|${REGISTRY}:${sha}|" "$manifest"
+  echo "  set $slot → $sha"
+}
+
+git_push() {
+  local msg=$1
+  cd "$INFRA_DIR"
+  git add -A
+  git commit -m "$msg"
+  git push
+  echo "  pushed: $msg"
+}
+
+wait_healthy() {
+  local slot=$1
+  local app="neuron"
+  [[ "$slot" != "prod" ]] && app="neuron-${slot}"
+  echo "  waiting for $slot to be healthy..."
+  local i=0
+  while [[ $i -lt 30 ]]; do
+    local ready
+    ready=$(KUBECONFIG="$KUBECONFIG" kubectl get pods -n neuron -l "app=${app}" \
+      --no-headers 2>/dev/null | grep "1/1" | wc -l)
+    [[ "$ready" -ge 1 ]] && echo "  $slot is healthy" && return 0
+    sleep 10
+    ((i++))
+  done
+  echo "  WARNING: $slot did not become healthy within 5 minutes"
+  return 1
+}
+
+# ── commands ──────────────────────────────────────────────────────────────────
+
+cmd_snapshot() {
+  local slot=${1:?Usage: snapshot <slot>}
+  echo "==> Snapshotting prod DB into $slot..."
+  KUBECONFIG="$KUBECONFIG" kubectl apply -f - << EOF
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: snapshot-prod-to-${slot}
+  namespace: neuron
+spec:
+  ttlSecondsAfterFinished: 300
+  template:
+    spec:
+      restartPolicy: Never
+      containers:
+        - name: snapshot
+          image: keinos/sqlite3:latest
+          command: ["/bin/sh", "-c"]
+          args:
+            - sqlite3 /src/neuron.db ".backup /dst/neuron.db" && echo "Done" && ls -lh /dst/neuron.db
+          volumeMounts:
+            - name: src
+              mountPath: /src
+              readOnly: true
+            - name: dst
+              mountPath: /dst
+      volumes:
+        - name: src
+          persistentVolumeClaim:
+            claimName: neuron-data
+        - name: dst
+          persistentVolumeClaim:
+            claimName: neuron-${slot}-data
+EOF
+  echo "  waiting for snapshot job..."
+  KUBECONFIG="$KUBECONFIG" kubectl wait job "snapshot-prod-to-${slot}" \
+    -n neuron --for=condition=complete --timeout=120s
+  echo "  snapshot complete"
+}
+
+cmd_deploy() {
+  local slot=${1:?Usage: deploy <slot> <sha>}
+  local sha=${2:?Usage: deploy <slot> <sha>}
+  echo "==> Deploying $sha to $slot..."
+  set_image "$slot" "$sha"
+  git_push "chore(neuron): deploy ${sha:0:12} to $slot"
+  wait_healthy "$slot"
+  echo "==> $slot is live at https://neuron-${slot}.neuralplatform.ai"
+}
+
+cmd_promote() {
+  local slot=${1:?Usage: promote <slot>}
+  local sha
+  sha=$(current_sha "$slot")
+  [[ -z "$sha" ]] && echo "ERROR: could not read SHA from $slot" && exit 1
+  echo "==> Promoting $slot ($sha) to prod..."
+  set_image "prod" "$sha"
+  # Also stamp the :stable registry tag (pull and re-tag)
+  git_push "chore(neuron): promote $slot (${sha:0:12}) to prod"
+  wait_healthy "prod"
+  echo "==> prod is now running $sha"
+  echo "==> prod endpoint: https://neuron.neuralplatform.ai"
+}
+
+cmd_rollback() {
+  local slot=${1:?Usage: rollback <slot> <sha>}
+  local sha=${2:?Usage: rollback <slot> <sha>}
+  echo "==> Rolling $slot back to $sha..."
+  set_image "$slot" "$sha"
+  git_push "chore(neuron): rollback $slot to ${sha:0:12}"
+  wait_healthy "$slot"
+  echo "==> $slot rolled back"
+}
+
+cmd_status() {
+  echo "==> Neuron slot status"
+  echo ""
+  for slot in prod alpha beta gamma; do
+    local app="neuron"
+    [[ "$slot" != "prod" ]] && app="neuron-${slot}"
+    local sha
+    sha=$(current_sha "$slot" 2>/dev/null || echo "unknown")
+    local ready
+    ready=$(KUBECONFIG="$KUBECONFIG" kubectl get pods -n neuron -l "app=${app}" \
+      --no-headers 2>/dev/null | grep "1/1" | wc -l | tr -d ' ')
+    printf "  %-8s  %s  pods_ready=%s\n" "$slot" "${sha:0:12}" "$ready"
+  done
+  echo ""
+  echo "  Endpoints:"
+  echo "    prod:  https://neuron.neuralplatform.ai"
+  echo "    alpha: https://neuron-alpha.neuralplatform.ai"
+  echo "    beta:  https://neuron-beta.neuralplatform.ai"
+  echo "    gamma: https://neuron-gamma.neuralplatform.ai"
+}
+
+# ── self-improvement loop (autonomous) ────────────────────────────────────────
+# When called with no args, prints the process for an autonomous agent to follow.
+cmd_loop_process() {
+  cat << 'PROCESS'
+NEURON SELF-IMPROVEMENT LOOP
+=============================
+
+Each iteration:
+
+1. IDENTIFY — find an improvement opportunity in the codebase.
+   Sources: Neuron backlog, observed failures, performance gaps, code review.
+
+2. IMPLEMENT — make the change on a git branch in neural-platform/neuron.
+   Commit message should describe the hypothesis being tested.
+
+3. CI — push the branch; CI runs lint/test/build and auto-deploys to alpha.
+   Wait for CI to complete and alpha pod to be healthy.
+
+4. SNAPSHOT — before the next experiment touches a slot, snapshot prod DB:
+     ./scripts/neuron-self-improve.sh snapshot <slot>
+   This ensures the experiment runs against real data without touching prod.
+
+5. EXPERIMENT — run the experiment against the alpha endpoint:
+     https://neuron-alpha.neuralplatform.ai
+   Use real workloads. Compare against prod or other slots.
+   Multiple experiments can run simultaneously across alpha/beta/gamma.
+
+6. EVALUATE — compare results. Criteria:
+   - Does it behave correctly?
+   - Does it perform better (latency, quality, fewer errors)?
+   - Does it handle edge cases prod handles?
+   Schema changes are acceptable — they become migrations against prod on promote.
+
+7. PROMOTE WINNER — if alpha wins:
+     ./scripts/neuron-self-improve.sh promote alpha
+   This updates prod's image SHA in git → Argo CD deploys → prod is updated.
+   The old prod SHA is still in the registry — rollback is always available.
+
+8. ROLLBACK — if something goes wrong:
+     ./scripts/neuron-self-improve.sh rollback prod <previous-sha>
+
+9. LOOP — go to step 1. Alpha is now free for the next experiment.
+
+RULES
+------
+- Never skip the snapshot step before running an experiment against a slot.
+- Prod data is never lost — slots only get copies.
+- Schema changes in experiment code → Alembic auto-migrates the snapshot DB.
+  On promote, the migration runs against prod. Always verify migrations are
+  additive/non-destructive before promoting.
+- All three experiment zones can run simultaneously with different hypotheses.
+- The registry retains every SHA — any version can be deployed to any slot.
+- This script handles git push; Argo CD handles k8s — never kubectl apply.
+
+PROCESS
+}
+
+# ── dispatch ──────────────────────────────────────────────────────────────────
+
+case "${1:-loop}" in
+  snapshot) cmd_snapshot "${@:2}" ;;
+  deploy)   cmd_deploy   "${@:2}" ;;
+  promote)  cmd_promote  "${@:2}" ;;
+  rollback) cmd_rollback "${@:2}" ;;
+  status)   cmd_status ;;
+  loop)     cmd_loop_process ;;
+  *)        echo "Usage: $0 snapshot|deploy|promote|rollback|status|loop" && exit 1 ;;
+esac