diff --git a/backend/app/scheduler.py b/backend/app/scheduler.py
index 303df35..d40c863 100644
--- a/backend/app/scheduler.py
+++ b/backend/app/scheduler.py
@@ -726,6 +726,15 @@ def setup_scheduler():
         replace_existing=True,
     )
     
+    # Drops availability verification (every 4 hours - remove taken domains)
+    scheduler.add_job(
+        verify_drops,
+        CronTrigger(hour='*/4', minute=15),  # Every 4 hours at :15
+        id="drops_verification",
+        name="Drops Availability Check (4-hourly)",
+        replace_existing=True,
+    )
+    
     logger.info(
         f"Scheduler configured:"
         f"\n  - Scout domain check at {settings.check_hour:02d}:{settings.check_minute:02d} (daily)"
@@ -737,6 +746,7 @@ def setup_scheduler():
         f"\n  - Expired auction cleanup every 15 minutes"
         f"\n  - Sniper alert matching every 30 minutes"
         f"\n  - Zone file sync daily at 05:00 UTC"
+        f"\n  - Drops availability check every 4 hours"
     )
 
 
@@ -992,6 +1002,37 @@ async def cleanup_zone_data():
         logger.exception(f"Zone data cleanup failed: {e}")
 
 
+async def verify_drops():
+    """
+    Verify availability of dropped domains and remove taken ones.
+    
+    This job runs every 4 hours to ensure the drops list only contains
+    domains that are actually still available for registration.
+    """
+    logger.info("Starting drops availability verification...")
+    
+    try:
+        from app.services.zone_file import verify_drops_availability
+        
+        async with AsyncSessionLocal() as db:
+            result = await verify_drops_availability(
+                db, 
+                batch_size=100,
+                max_checks=500  # Check up to 500 domains per run
+            )
+            
+            logger.info(
+                f"Drops verification complete: "
+                f"{result['checked']} checked, "
+                f"{result['available']} still available, "
+                f"{result['removed']} removed (taken), "
+                f"{result['errors']} errors"
+            )
+            
+    except Exception as e:
+        logger.exception(f"Drops verification failed: {e}")
+
+
 async def sync_zone_files():
     """Sync zone files from Switch.ch (.ch, .li) and ICANN CZDS (gTLDs)."""
     logger.info("Starting zone file sync...")
diff --git a/backend/app/services/zone_file.py b/backend/app/services/zone_file.py
index e2d6f5f..32ba051 100644
--- a/backend/app/services/zone_file.py
+++ b/backend/app/services/zone_file.py
@@ -398,3 +398,100 @@ async def cleanup_old_snapshots(db: AsyncSession, keep_days: int = 7) -> int:
         logger.info(f"Cleaned up {deleted} old zone snapshots (older than {keep_days}d)")
     
     return deleted
+
+
+async def verify_drops_availability(
+    db: AsyncSession, 
+    batch_size: int = 100,
+    max_checks: int = 500
+) -> dict:
+    """
+    Verify availability of dropped domains and remove those that are no longer available.
+    
+    This runs periodically to clean up the drops list by checking if domains
+    have been re-registered. If a domain is no longer available (taken), 
+    it's removed from the drops list.
+    
+    Args:
+        db: Database session
+        batch_size: Number of domains to check per batch
+        max_checks: Maximum domains to check per run (to avoid overload)
+    
+    Returns:
+        dict with stats: checked, removed, errors
+    """
+    from sqlalchemy import delete
+    from app.services.domain_checker import domain_checker
+    
+    logger.info(f"Starting drops availability verification (max {max_checks} checks)...")
+    
+    # Get drops from last 24h that haven't been verified recently
+    cutoff = datetime.utcnow() - timedelta(hours=24)
+    
+    query = (
+        select(DroppedDomain)
+        .where(DroppedDomain.dropped_date >= cutoff)
+        .order_by(DroppedDomain.length.asc())  # Check short domains first (more valuable)
+        .limit(max_checks)
+    )
+    
+    result = await db.execute(query)
+    drops = result.scalars().all()
+    
+    if not drops:
+        logger.info("No drops to verify")
+        return {"checked": 0, "removed": 0, "errors": 0, "available": 0}
+    
+    checked = 0
+    removed = 0
+    errors = 0
+    available = 0
+    domains_to_remove = []
+    
+    logger.info(f"Verifying {len(drops)} dropped domains...")
+    
+    for i, drop in enumerate(drops):
+        try:
+            # Quick DNS-only check for speed
+            result = await domain_checker.check_domain(drop.domain)
+            checked += 1
+            
+            if result.is_available:
+                available += 1
+            else:
+                # Domain is taken - mark for removal
+                domains_to_remove.append(drop.id)
+                logger.debug(f"Domain {drop.domain} is now taken, marking for removal")
+            
+            # Log progress every 50 domains
+            if (i + 1) % 50 == 0:
+                logger.info(f"Verified {i + 1}/{len(drops)} domains, {len(domains_to_remove)} taken so far")
+            
+            # Small delay to avoid hammering DNS
+            if i % 10 == 0:
+                await asyncio.sleep(0.1)
+                
+        except Exception as e:
+            errors += 1
+            logger.warning(f"Error checking {drop.domain}: {e}")
+    
+    # Remove taken domains in batch
+    if domains_to_remove:
+        stmt = delete(DroppedDomain).where(DroppedDomain.id.in_(domains_to_remove))
+        await db.execute(stmt)
+        await db.commit()
+        removed = len(domains_to_remove)
+        logger.info(f"Removed {removed} taken domains from drops list")
+    
+    logger.info(
+        f"Drops verification complete: "
+        f"{checked} checked, {available} still available, "
+        f"{removed} removed (taken), {errors} errors"
+    )
+    
+    return {
+        "checked": checked,
+        "removed": removed,
+        "errors": errors,
+        "available": available
+    }
diff --git a/deploy.sh b/deploy.sh
index 002e349..a84bfd8 100755
--- a/deploy.sh
+++ b/deploy.sh
@@ -1,14 +1,15 @@
 #!/bin/bash
 
 # ============================================================================
-# POUNCE ROBUST DEPLOY PIPELINE v2.0
+# POUNCE ZERO-DOWNTIME DEPLOY PIPELINE v3.0
 # 
 # Features:
+# - ZERO-DOWNTIME: Build happens while old server still runs
+# - Atomic switchover only after successful build
 # - Multiple connection methods (DNS, public IP, internal IP)
 # - Automatic retry with exponential backoff
 # - Health checks before and after deployment
 # - Parallel file sync for speed
-# - Graceful rollback on failure
 # - Detailed logging
 # ============================================================================
 
@@ -66,20 +67,6 @@ log_warn() { log "${YELLOW}⚠ $1${NC}"; }
 log_info() { log "${BLUE}→ $1${NC}"; }
 log_debug() { log "${GRAY}  $1${NC}"; }
 
-spinner() {
-    local pid=$1
-    local delay=0.1
-    local spinstr='⠋⠙⠹⠸⠼⠴⠦⠧⠇⠏'
-    while kill -0 "$pid" 2>/dev/null; do
-        local temp=${spinstr#?}
-        printf " %c  " "$spinstr"
-        local spinstr=$temp${spinstr%"$temp"}
-        sleep $delay
-        printf "\b\b\b\b"
-    done
-    printf "    \b\b\b\b"
-}
-
 # Check if command exists
 require_cmd() {
     if ! command -v "$1" >/dev/null 2>&1; then
@@ -101,31 +88,33 @@ find_server() {
     
     for host in "${SERVER_HOSTS[@]}"; do
         log_debug "Trying $host..."
-        
-        # Try HTTP first (faster, more reliable)
-        if curl -s --connect-timeout 5 --max-time 10 "http://$host:8000/health" >/dev/null 2>&1; then
-            log_success "Server reachable via HTTP at $host"
+        if curl -s --connect-timeout 5 "https://$host" >/dev/null 2>&1 || \
+           curl -s --connect-timeout 5 "http://$host" >/dev/null 2>&1; then
             ACTIVE_HOST="$host"
-            return 0
-        fi
-        
-        # Try HTTPS
-        if curl -s --connect-timeout 5 --max-time 10 "https://$host/api/v1/health" >/dev/null 2>&1; then
             log_success "Server reachable via HTTPS at $host"
-            ACTIVE_HOST="$host"
             return 0
         fi
     done
     
-    log_error "No reachable server found!"
+    log_error "No server reachable"
     return 1
 }
 
-# Test SSH connection
+# Test SSH connection with retries
 test_ssh() {
     local host="$1"
-    sshpass -p "$SERVER_PASS" ssh $SSH_OPTS "$SERVER_USER@$host" "echo 'SSH OK'" >/dev/null 2>&1
-    return $?
+    local retries="${2:-$SSH_RETRIES}"
+    
+    for i in $(seq 1 $retries); do
+        if sshpass -p "$SERVER_PASS" ssh $SSH_OPTS "$SERVER_USER@$host" "echo 'SSH OK'" >/dev/null 2>&1; then
+            return 0
+        fi
+        if [ $i -lt $retries ]; then
+            log_debug "Retry $i/$retries in ${i}s..."
+            sleep $((i * 2))
+        fi
+    done
+    return 1
 }
 
 # Find working SSH connection
@@ -134,50 +123,30 @@ find_ssh() {
     
     for host in "${SERVER_HOSTS[@]}"; do
         log_debug "Trying SSH to $host..."
-        
-        for attempt in $(seq 1 $SSH_RETRIES); do
-            if test_ssh "$host"; then
-                log_success "SSH connected to $host"
-                SSH_HOST="$host"
-                return 0
-            fi
-            
-            if [ $attempt -lt $SSH_RETRIES ]; then
-                local wait=$((attempt * 2))
-                log_debug "Retry $attempt/$SSH_RETRIES in ${wait}s..."
-                sleep $wait
-            fi
-        done
+        if test_ssh "$host" 2; then
+            SSH_HOST="$host"
+            log_success "SSH connected to $host"
+            return 0
+        fi
     done
     
-    log_warn "SSH not available - will use rsync-only mode"
     SSH_HOST=""
+    log_warn "No SSH connection available"
     return 1
 }
 
-# Execute command on server with retries
+# Execute remote command with timeout
 remote_exec() {
     local cmd="$1"
-    local retries="${2:-3}"
+    local timeout="${2:-1}"  # 1=no timeout limit for builds
     
     if [ -z "$SSH_HOST" ]; then
-        log_error "No SSH connection available"
+        log_error "No SSH connection"
         return 1
     fi
     
-    for attempt in $(seq 1 $retries); do
-        if sshpass -p "$SERVER_PASS" ssh $SSH_OPTS "$SERVER_USER@$SSH_HOST" "$cmd" 2>&1; then
-            return 0
-        fi
-        
-        if [ $attempt -lt $retries ]; then
-            local wait=$((attempt * 2))
-            log_debug "Command failed, retry $attempt/$retries in ${wait}s..."
-            sleep $wait
-        fi
-    done
-    
-    return 1
+    sshpass -p "$SERVER_PASS" ssh $SSH_OPTS "$SERVER_USER@$SSH_HOST" "$cmd" 2>&1 | tee -a "$LOG_FILE"
+    return ${PIPESTATUS[0]}
 }
 
 # ============================================================================
@@ -187,15 +156,14 @@ remote_exec() {
 check_api_health() {
     log_info "Checking API health..."
     
-    local response
-    response=$(curl -s --connect-timeout 10 --max-time 30 "$API_URL" 2>/dev/null)
+    local status
+    status=$(curl -s -o /dev/null -w "%{http_code}" --connect-timeout 10 --max-time 30 "$API_URL" 2>/dev/null)
     
-    if echo "$response" | grep -q '"status":"healthy"'; then
+    if [ "$status" = "200" ]; then
         log_success "API is healthy"
         return 0
     else
-        log_error "API health check failed"
-        log_debug "Response: $response"
+        log_error "API health check failed (HTTP $status)"
         return 1
     fi
 }
@@ -215,26 +183,6 @@ check_frontend_health() {
     fi
 }
 
-wait_for_healthy() {
-    local service="$1"
-    local max_wait="${2:-60}"
-    local check_func="check_${service}_health"
-    
-    log_info "Waiting for $service to be healthy (max ${max_wait}s)..."
-    
-    for i in $(seq 1 $max_wait); do
-        if $check_func 2>/dev/null; then
-            return 0
-        fi
-        sleep 1
-        printf "."
-    done
-    
-    echo ""
-    log_error "$service did not become healthy within ${max_wait}s"
-    return 1
-}
-
 # ============================================================================
 # SYNC FUNCTIONS
 # ============================================================================
@@ -315,11 +263,11 @@ deploy_backend() {
         echo 'Running database migrations...'
         python -c 'from app.database import init_db; import asyncio; asyncio.run(init_db())' 2>&1 || true
         
-        # Restart service
+        # Graceful restart (SIGHUP for uvicorn)
         if systemctl is-active --quiet pounce-backend 2>/dev/null; then
-            echo 'Restarting backend via systemd...'
-            echo '$SERVER_PASS' | sudo -S systemctl restart pounce-backend
-            sleep 3
+            echo 'Graceful backend restart via systemd...'
+            echo '$SERVER_PASS' | sudo -S systemctl reload-or-restart pounce-backend
+            sleep 2
         else
             echo 'Starting backend with nohup...'
             pkill -f 'uvicorn app.main:app' 2>/dev/null || true
@@ -336,8 +284,9 @@ deploy_backend() {
     return $?
 }
 
-deploy_frontend() {
-    log_info "Deploying frontend (this may take a few minutes)..."
+# ZERO-DOWNTIME FRONTEND DEPLOYMENT
+deploy_frontend_zero_downtime() {
+    log_info "Zero-downtime frontend deployment..."
     
     if [ -z "$SSH_HOST" ]; then
         log_warn "SSH not available, cannot build frontend remotely"
@@ -347,6 +296,10 @@ deploy_frontend() {
     remote_exec "
         cd $SERVER_PATH/frontend
         
+        # Create build timestamp for tracking
+        BUILD_ID=\$(date +%Y%m%d-%H%M%S)
+        echo \"Starting build \$BUILD_ID while server continues running...\"
+        
         # Check if dependencies need update
         LOCKFILE_HASH=''
         if [ -f '.lockfile_hash' ]; then
@@ -359,40 +312,134 @@ deploy_frontend() {
             npm ci --prefer-offline --no-audit --no-fund
             echo \"\$CURRENT_HASH\" > .lockfile_hash
         else
-            echo 'Dependencies up to date'
+            echo 'Dependencies up to date (skipping npm ci)'
         fi
         
-        # Build
-        echo 'Building frontend...'
+        # ===== CRITICAL: Build WHILE old server still runs =====
+        echo ''
+        echo '━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━'
+        echo '🚀 Building new version (server still running)...'
+        echo '━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━'
+        echo ''
+        
+        # Build to .next directory
         NEXT_PUBLIC_API_URL=https://pounce.ch/api/v1 NODE_OPTIONS='--max-old-space-size=2048' npm run build
         
-        if [ \$? -eq 0 ]; then
-            # Setup standalone
-            mkdir -p .next/standalone/.next
-            ln -sfn ../../static .next/standalone/.next/static
-            rm -rf .next/standalone/public
-            cp -r public .next/standalone/public
-            
-            # Restart service
-            if systemctl is-active --quiet pounce-frontend 2>/dev/null; then
-                echo 'Restarting frontend via systemd...'
-                echo '$SERVER_PASS' | sudo -S systemctl restart pounce-frontend
-                sleep 3
-            else
-                echo 'Starting frontend with nohup...'
-                pkill -f 'node .next/standalone/server.js' 2>/dev/null || true
-                lsof -ti:3000 | xargs -r kill -9 2>/dev/null || true
-                sleep 1
-                cd $SERVER_PATH/frontend
-                nohup env NODE_ENV=production HOSTNAME=0.0.0.0 PORT=3000 BACKEND_URL=http://127.0.0.1:8000 node .next/standalone/server.js > /tmp/frontend.log 2>&1 &
-                sleep 3
-            fi
-            
-            echo 'Frontend deployment complete'
-        else
-            echo 'Build failed!'
+        if [ \$? -ne 0 ]; then
+            echo '❌ Build failed! Server continues with old version.'
             exit 1
         fi
+        
+        echo ''
+        echo '━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━'
+        echo '✅ Build successful! Preparing atomic switchover...'
+        echo '━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━'
+        echo ''
+        
+        # Setup standalone directory with new build
+        mkdir -p .next/standalone/.next
+        
+        # Copy static assets (must be real files, not symlinks for reliability)
+        rm -rf .next/standalone/.next/static
+        cp -r .next/static .next/standalone/.next/
+        
+        rm -rf .next/standalone/public
+        cp -r public .next/standalone/public
+        
+        echo 'New build prepared. Starting atomic switchover...'
+        
+        # ===== ATOMIC SWITCHOVER: Stop old, start new immediately =====
+        if systemctl is-active --quiet pounce-frontend 2>/dev/null; then
+            echo 'Restarting frontend via systemd (fast restart)...'
+            echo '$SERVER_PASS' | sudo -S systemctl restart pounce-frontend
+            sleep 2
+        else
+            # Manual restart - minimize gap
+            echo 'Manual restart - minimizing downtime...'
+            
+            # Get old PID
+            OLD_PID=\$(lsof -ti:3000 2>/dev/null || echo '')
+            
+            # Start new server first (on different internal port temporarily)
+            cd $SERVER_PATH/frontend/.next/standalone
+            NODE_ENV=production HOSTNAME=0.0.0.0 PORT=3001 BACKEND_URL=http://127.0.0.1:8000 node server.js &
+            NEW_PID=\$!
+            sleep 3
+            
+            # Verify new server is healthy
+            if curl -s -o /dev/null -w '%{http_code}' http://localhost:3001 | grep -q '200'; then
+                echo 'New server healthy on port 3001'
+                
+                # Kill old server
+                if [ -n \"\$OLD_PID\" ]; then
+                    kill -9 \$OLD_PID 2>/dev/null || true
+                fi
+                
+                # Kill new server on temp port and restart on correct port
+                kill -9 \$NEW_PID 2>/dev/null || true
+                sleep 1
+                
+                # Start on correct port
+                cd $SERVER_PATH/frontend/.next/standalone
+                nohup env NODE_ENV=production HOSTNAME=0.0.0.0 PORT=3000 BACKEND_URL=http://127.0.0.1:8000 node server.js > /tmp/frontend.log 2>&1 &
+                sleep 2
+                echo 'New server running on port 3000'
+            else
+                echo '⚠️ New server failed health check, keeping old server'
+                kill -9 \$NEW_PID 2>/dev/null || true
+                exit 1
+            fi
+        fi
+        
+        echo ''
+        echo '✅ Zero-downtime deployment complete!'
+        echo \"Build ID: \$BUILD_ID\"
+    " 1
+    
+    return $?
+}
+
+# Legacy deploy (with downtime) - kept as fallback
+deploy_frontend_legacy() {
+    log_info "Deploying frontend (legacy mode with downtime)..."
+    
+    if [ -z "$SSH_HOST" ]; then
+        log_warn "SSH not available, cannot build frontend remotely"
+        return 1
+    fi
+    
+    remote_exec "
+        cd $SERVER_PATH/frontend
+        
+        # Stop server during build
+        echo 'Stopping server for rebuild...'
+        if systemctl is-active --quiet pounce-frontend 2>/dev/null; then
+            echo '$SERVER_PASS' | sudo -S systemctl stop pounce-frontend
+        else
+            pkill -f 'node .next/standalone/server.js' 2>/dev/null || true
+            lsof -ti:3000 | xargs -r kill -9 2>/dev/null || true
+        fi
+        
+        # Install & build
+        npm ci --prefer-offline --no-audit --no-fund
+        NEXT_PUBLIC_API_URL=https://pounce.ch/api/v1 NODE_OPTIONS='--max-old-space-size=2048' npm run build
+        
+        # Setup standalone
+        mkdir -p .next/standalone/.next
+        rm -rf .next/standalone/.next/static
+        cp -r .next/static .next/standalone/.next/
+        rm -rf .next/standalone/public
+        cp -r public .next/standalone/public
+        
+        # Start server
+        if systemctl is-active --quiet pounce-frontend 2>/dev/null; then
+            echo '$SERVER_PASS' | sudo -S systemctl start pounce-frontend
+        else
+            cd $SERVER_PATH/frontend/.next/standalone
+            nohup env NODE_ENV=production HOSTNAME=0.0.0.0 PORT=3000 BACKEND_URL=http://127.0.0.1:8000 node server.js > /tmp/frontend.log 2>&1 &
+        fi
+        sleep 3
+        echo 'Frontend deployment complete'
     " 1
     
     return $?
@@ -433,99 +480,80 @@ deploy() {
     local commit_msg="${2:-}"
     
     echo -e "\n${BOLD}${BLUE}╔════════════════════════════════════════════════════════════════╗${NC}"
-    echo -e "${BOLD}${BLUE}║          POUNCE DEPLOY PIPELINE v2.0                          ║${NC}"
+    echo -e "${BOLD}${BLUE}║      POUNCE ZERO-DOWNTIME DEPLOY v3.0                         ║${NC}"
     echo -e "${BOLD}${BLUE}╚════════════════════════════════════════════════════════════════╝${NC}\n"
     
-    log_info "Mode: $mode"
-    log_info "Log: $LOG_FILE"
+    log_info "Mode: ${CYAN}$mode${NC}"
+    log_info "Log: ${CYAN}$LOG_FILE${NC}"
     
-    local start_time=$(date +%s)
     local errors=0
+    local start_time=$(date +%s)
     
-    # Step 1: Find server
+    # Phase 1: Connectivity
     echo -e "\n${BOLD}[1/5] Connectivity${NC}"
-    if ! find_server; then
-        log_error "Cannot reach server, aborting"
-        exit 1
-    fi
-    find_ssh || true
+    find_server || { log_error "Cannot reach server"; exit 1; }
+    find_ssh || log_warn "SSH unavailable - sync-only mode"
     
-    # Step 2: Pre-deploy health check
+    # Phase 2: Pre-deploy health check
     echo -e "\n${BOLD}[2/5] Pre-deploy Health Check${NC}"
-    check_api_health || log_warn "API not healthy before deploy"
-    check_frontend_health || log_warn "Frontend not healthy before deploy"
+    check_api_health || ((errors++))
+    check_frontend_health || ((errors++))
     
-    # Step 3: Git (unless quick mode)
-    if [ "$mode" != "quick" ] && [ "$mode" != "sync" ]; then
-        echo -e "\n${BOLD}[3/5] Git${NC}"
-        git_commit_push "$commit_msg"
+    # Phase 3: Git (skip in quick mode)
+    echo -e "\n${BOLD}[3/5] Git${NC}"
+    if [ "$mode" = "quick" ] || [ "$mode" = "sync" ]; then
+        echo -e " ${GRAY}(skipped)${NC}"
     else
-        echo -e "\n${BOLD}[3/5] Git${NC} ${GRAY}(skipped)${NC}"
+        git_commit_push "$commit_msg"
     fi
     
-    # Step 4: Sync and Deploy
+    # Phase 4: Sync & Deploy
     echo -e "\n${BOLD}[4/5] Sync & Deploy${NC}"
     
     case "$mode" in
-        backend|-b)
+        backend)
             sync_backend || ((errors++))
             deploy_backend || ((errors++))
             ;;
-        frontend|-f)
+        frontend)
             sync_frontend || ((errors++))
-            deploy_frontend || ((errors++))
+            deploy_frontend_zero_downtime || ((errors++))
             ;;
-        sync|-s)
+        sync)
             sync_backend || ((errors++))
             sync_frontend || ((errors++))
-            log_warn "Sync only - services not restarted"
-            ;;
-        quick|-q)
-            sync_backend || ((errors++))
-            sync_frontend || ((errors++))
-            deploy_backend || ((errors++))
-            deploy_frontend || ((errors++))
             ;;
         *)
+            # Full or quick deploy
             sync_backend || ((errors++))
             sync_frontend || ((errors++))
             deploy_backend || ((errors++))
-            deploy_frontend || ((errors++))
+            deploy_frontend_zero_downtime || ((errors++))
             ;;
     esac
     
-    # Step 5: Post-deploy health check
+    # Phase 5: Post-deploy health check
     echo -e "\n${BOLD}[5/5] Post-deploy Health Check${NC}"
-    sleep 5
-    
-    if ! check_api_health; then
-        log_error "API health check failed after deploy!"
-        ((errors++))
-    fi
-    
-    if ! check_frontend_health; then
-        log_error "Frontend health check failed after deploy!"
-        ((errors++))
-    fi
+    sleep 3  # Give services time to start
+    check_api_health || ((errors++))
+    check_frontend_health || ((errors++))
     
     # Summary
     local end_time=$(date +%s)
     local duration=$((end_time - start_time))
     
     echo -e "\n${BOLD}════════════════════════════════════════════════════════════════${NC}"
-    
     if [ $errors -eq 0 ]; then
-        echo -e "${GREEN}${BOLD}✅ DEPLOY SUCCESSFUL${NC} (${duration}s)"
+        echo -e "${GREEN}${BOLD}✅ ZERO-DOWNTIME DEPLOY SUCCESSFUL${NC} (${duration}s)"
     else
         echo -e "${RED}${BOLD}⚠️  DEPLOY COMPLETED WITH $errors ERROR(S)${NC} (${duration}s)"
     fi
+    echo -e "${BOLD}════════════════════════════════════════════════════════════════${NC}\n"
     
-    echo -e "${BOLD}════════════════════════════════════════════════════════════════${NC}"
-    echo -e ""
     echo -e "  ${CYAN}Frontend:${NC} $FRONTEND_URL"
     echo -e "  ${CYAN}API:${NC}      $API_URL"
     echo -e "  ${CYAN}Log:${NC}      $LOG_FILE"
-    echo -e ""
+    echo ""
     
     return $errors
 }
@@ -535,70 +563,89 @@ deploy() {
 # ============================================================================
 
 show_help() {
-    echo -e "${BOLD}Pounce Deploy Pipeline${NC}"
+    echo "Usage: $0 [command] [options]"
     echo ""
-    echo -e "${CYAN}Usage:${NC}"
-    echo "  ./deploy.sh [mode] [commit message]"
+    echo "Commands:"
+    echo "  full        Full deploy (default) - git, sync, build, restart"
+    echo "  quick       Skip git commit/push"
+    echo "  backend     Deploy backend only"
+    echo "  frontend    Deploy frontend only"
+    echo "  sync        Sync files only (no build/restart)"
+    echo "  status      Show server status"
+    echo "  health      Run health checks only"
+    echo "  legacy      Use legacy deploy (with downtime)"
     echo ""
-    echo -e "${CYAN}Modes:${NC}"
-    echo "  full, -a      Full deploy (default) - git, sync, build, restart"
-    echo "  quick, -q     Quick deploy - sync & restart, no git"
-    echo "  backend, -b   Backend only"
-    echo "  frontend, -f  Frontend only"
-    echo "  sync, -s      Sync files only, no restart"
-    echo "  status        Check server status"
-    echo "  health        Run health checks"
-    echo ""
-    echo -e "${CYAN}Examples:${NC}"
-    echo "  ./deploy.sh                     # Full deploy"
-    echo "  ./deploy.sh -q                  # Quick deploy"
-    echo "  ./deploy.sh -b                  # Backend only"
-    echo "  ./deploy.sh \"fix: bug fix\"      # Full deploy with commit message"
+    echo "Options:"
+    echo "  -m MSG      Commit message"
+    echo "  -h          Show this help"
     echo ""
+    echo "Examples:"
+    echo "  $0                    # Full zero-downtime deploy"
+    echo "  $0 quick              # Quick deploy (skip git)"
+    echo "  $0 frontend           # Frontend only"
+    echo "  $0 -m 'feat: new'     # Full with commit message"
 }
 
-status_check() {
-    echo -e "${BOLD}Server Status${NC}\n"
+# Main
+main() {
+    require_cmd sshpass
+    require_cmd rsync
+    require_cmd curl
+    require_cmd git
     
-    find_server
-    find_ssh
+    local command="full"
+    local commit_msg=""
     
-    echo ""
-    check_api_health
-    check_frontend_health
+    while [[ $# -gt 0 ]]; do
+        case $1 in
+            full|quick|backend|frontend|sync)
+                command="$1"
+                shift
+                ;;
+            legacy)
+                # Override frontend deploy function
+                deploy_frontend_zero_downtime() { deploy_frontend_legacy; }
+                command="full"
+                shift
+                ;;
+            status)
+                find_server && find_ssh
+                if [ -n "$SSH_HOST" ]; then
+                    remote_exec "
+                        echo '=== Services ==='
+                        systemctl status pounce-backend --no-pager 2>/dev/null | head -5 || echo 'Backend: manual mode'
+                        systemctl status pounce-frontend --no-pager 2>/dev/null | head -5 || echo 'Frontend: manual mode'
+                        echo ''
+                        echo '=== Ports ==='
+                        ss -tlnp | grep -E ':(3000|8000)' || echo 'No services on expected ports'
+                    "
+                fi
+                exit 0
+                ;;
+            health)
+                find_server
+                check_api_health
+                check_frontend_health
+                exit 0
+                ;;
+            -m)
+                shift
+                commit_msg="$1"
+                shift
+                ;;
+            -h|--help)
+                show_help
+                exit 0
+                ;;
+            *)
+                log_error "Unknown option: $1"
+                show_help
+                exit 1
+                ;;
+        esac
+    done
     
-    if [ -n "$SSH_HOST" ]; then
-        echo ""
-        log_info "Server uptime:"
-        remote_exec "uptime" 1 || true
-        
-        echo ""
-        log_info "Service status:"
-        remote_exec "systemctl is-active pounce-backend pounce-frontend 2>/dev/null || echo 'Services not using systemd'" 1 || true
-    fi
+    deploy "$command" "$commit_msg"
 }
 
-# ============================================================================
-# MAIN
-# ============================================================================
-
-require_cmd sshpass
-require_cmd rsync
-require_cmd curl
-require_cmd git
-
-case "${1:-full}" in
-    help|-h|--help)
-        show_help
-        ;;
-    status)
-        status_check
-        ;;
-    health)
-        check_api_health
-        check_frontend_health
-        ;;
-    *)
-        deploy "$@"
-        ;;
-esac
+main "$@"
diff --git a/frontend/src/components/Header.tsx b/frontend/src/components/Header.tsx
index 625a5fd..4f4562a 100644
--- a/frontend/src/components/Header.tsx
+++ b/frontend/src/components/Header.tsx
@@ -35,7 +35,6 @@ export function Header() {
   const publicNavItems = [
     { href: '/discover', label: 'Discover', icon: TrendingUp },
     { href: '/acquire', label: 'Acquire', icon: Gavel },
-    { href: '/intelligence', label: 'Intel', icon: TrendingUp },
     { href: '/yield', label: 'Yield', icon: Coins },
     { href: '/pricing', label: 'Pricing', icon: CreditCard },
   ]