diff --git a/client/internal/auth/device_flow.go b/client/internal/auth/device_flow.go
index 9dec7cf53..3592e589d 100644
--- a/client/internal/auth/device_flow.go
+++ b/client/internal/auth/device_flow.go
@@ -304,6 +304,16 @@ func (d *DeviceAuthorizationFlow) WaitToken(ctx context.Context, info AuthFlowIn
 				return TokenInfo{}, fmt.Errorf("validate access token failed with error: %v", err)
 			}
 
+			// Same as the PKCE flow: the account the token belongs to is what
+			// callers store to send back as the login_hint. Without it a client
+			// driven through the device flow — Android TV and tvOS — never binds
+			// an account to its profile and every later login goes out blind.
+			if email, err := parseEmailFromIDToken(tokenInfo.IDToken); err != nil {
+				log.Warnf("failed to parse email from ID token: %v", err)
+			} else {
+				tokenInfo.Email = email
+			}
+
 			log.Infof("device flow: user authorization confirmed after %d polls in %s", polls, time.Since(start).Round(time.Second))
 			return tokenInfo, err
 		}
diff --git a/client/ios/NetBirdSDK/login.go b/client/ios/NetBirdSDK/login.go
index 42a575359..cf7aa6730 100644
--- a/client/ios/NetBirdSDK/login.go
+++ b/client/ios/NetBirdSDK/login.go
@@ -11,6 +11,7 @@ import (
 
 	"github.com/netbirdio/netbird/client/internal/auth"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
+	"github.com/netbirdio/netbird/client/mobile"
 	"github.com/netbirdio/netbird/client/system"
 )
 
@@ -284,12 +285,14 @@ func (a *Auth) login(urlOpener URLOpener, forceDeviceAuth bool, deviceName strin
 	}
 
 	jwtToken := ""
+	email := ""
 	if needsLogin {
 		tokenInfo, err := a.foregroundGetTokenInfo(authClient, urlOpener, forceDeviceAuth)
 		if err != nil {
 			return fmt.Errorf("interactive sso login failed: %v", err)
 		}
 		jwtToken = tokenInfo.GetTokenToUse()
+		email = tokenInfo.Email
 	}
 
 	err, isAuthError := authClient.Login(ctx, "", jwtToken)
@@ -301,6 +304,14 @@ func (a *Auth) login(urlOpener URLOpener, forceDeviceAuth bool, deviceName strin
 		return fmt.Errorf("login failed: %v", err)
 	}
 
+	// Stored after Login, not before: a rejected token must not leave a hint
+	// pointing at an account that cannot be used.
+	if email != "" && a.cfgPath != "" {
+		if err := mobile.WriteProfileEmail(a.cfgPath, email); err != nil {
+			log.Warnf("failed to store profile account email: %v", err)
+		}
+	}
+
 	// Save the config before notifying success to ensure persistence completes
 	// before the callback potentially triggers teardown on the Swift side.
 	// Note: This differs from Android which doesn't save config after login.
@@ -320,10 +331,24 @@ func (a *Auth) login(urlOpener URLOpener, forceDeviceAuth bool, deviceName strin
 	return nil
 }
 
+// profileLoginHint returns the stored account email for the profile at cfgPath,
+// so a re-login targets the account the profile already belongs to instead of
+// whatever session the shared browser cookie jar happens to hold.
+//
+// An empty hint is deliberate, not a fallback: a fresh profile leaves the
+// choice to the IdP. Switching accounts is done by switching or removing
+// profiles, not by logging out — logout keeps the email.
+func profileLoginHint(cfgPath string) string {
+	if cfgPath == "" {
+		return ""
+	}
+	return mobile.ReadProfileEmail(cfgPath)
+}
+
 const authInfoRequestTimeout = 30 * time.Second
 
 func (a *Auth) foregroundGetTokenInfo(authClient *auth.Auth, urlOpener URLOpener, forceDeviceAuth bool) (*auth.TokenInfo, error) {
-	oAuthFlow, err := authClient.GetOAuthFlow(a.ctx, forceDeviceAuth, "")
+	oAuthFlow, err := authClient.GetOAuthFlow(a.ctx, forceDeviceAuth, profileLoginHint(a.cfgPath))
 	if err != nil {
 		return nil, fmt.Errorf("failed to get OAuth flow: %v", err)
 	}
diff --git a/client/mobile/profile_state.go b/client/mobile/profile_state.go
index bb983ec1d..ad05801f8 100644
--- a/client/mobile/profile_state.go
+++ b/client/mobile/profile_state.go
@@ -78,8 +78,14 @@ func WriteProfileEmail(configPath string, email string) error {
 		return fmt.Errorf("resolve profile account path: %w", err)
 	}
 
+	// DirectWriteJson, not the atomic writers: those create a temp file and
+	// rename it over the target, which the tvOS App Group sandbox blocks. It is
+	// the same reason the config next to this file goes through
+	// DirectWriteOutConfig. The file is rewritten whole from one key, so losing
+	// atomicity costs nothing beyond a torn write on a crash mid-write, which
+	// reads back as "no email" and is recovered by the next login.
 	state := profilemanager.ProfileState{Email: email}
-	if err := util.WriteJsonWithRestrictedPermission(context.Background(), accountPath, state); err != nil {
+	if err := util.DirectWriteJson(context.Background(), accountPath, state); err != nil {
 		return fmt.Errorf("write profile account: %w", err)
 	}
 
diff --git a/util/file.go b/util/file.go
index 926904f9f..52eb91c0f 100644
--- a/util/file.go
+++ b/util/file.go
@@ -56,9 +56,9 @@ func WriteJson(ctx context.Context, file string, obj interface{}) error {
 }
 
 // DirectWriteJson writes JSON config object to a file creating parent directories if required without creating a temporary file
-func DirectWriteJson(ctx context.Context, file string, obj interface{}) error {
+func DirectWriteJson(ctx context.Context, file string, obj interface{}) (err error) {
 
-	_, _, err := prepareConfigFileDir(file)
+	_, _, err = prepareConfigFileDir(file)
 	if err != nil {
 		return err
 	}
@@ -68,11 +68,24 @@ func DirectWriteJson(ctx context.Context, file string, obj interface{}) error {
 		return err
 	}
 
+	// Named return so a failed Close is reported rather than logged and
+	// swallowed: the write is only durable once the file closes cleanly, and a
+	// caller told "written" would carry on with data that never landed.
 	defer func() {
-		err = targetFile.Close()
-		if err != nil {
-			log.Errorf("failed to close file %s: %v", file, err)
+		cerr := targetFile.Close()
+		if cerr == nil {
+			return
 		}
+		if err == nil {
+			// Returned, not logged: the caller reports it once.
+			err = cerr
+			return
+		}
+		// The body already failed and that error is the one the caller gets, so
+		// it is the one that explains the failure. This is then the only place
+		// the close failure can surface — at debug, per the logging rules for
+		// close errors on writes.
+		log.Debugf("failed to close file %s after %v: %v", file, err, cerr)
 	}()
 
 	// make it pretty