• Agent upgrade

  • Optimized Helm Agent v1.8.0 performance to reduce CPU and memory resource consumption up to 60% without impacting data collection.
    Upgrade your Kubernetes agent to the latest version with this command.

helm repo update middleware-labs
helm upgrade mw-agent middleware-labs/mw-kube-agent-v3 \ --version 1.8.0 \ --reset-then-reuse-values \ -n mw-agent-ns

• Integration enhancements

  • Added Kubernetes auto-instrumentation Agent to support for GCP MySQL monitoring.

  • GCP integrations now support folder-level data polling.
    

• Product improvements

  • Alert : Webhook notifications now support custom headers and payloads.

  • RUM : Added role-based access control (RBAC) for the RUM module.

  • OpsAI : Improved OpsAI performance for faster responses and more accurate chart citations.

  • Dashboards : Enhanced Dashboard Builder usability with improved charts, legends, and layouts.

Integrations

Auto-Instrumentation for ECS
Configure AWS ECS task definitions and automatically instrument applications running inside ECS with a single click.

Prometheus Integration
Added Basic Authentication support for scraping password-protected Prometheus endpoints.

Kafka Integration
Added Kubernetes support to monitor production Kafka environments.

Log Anomalies and Audit logs

Automatically detect log anomalies such as count spikes, behavioral drift, and inactive patterns.

Drill down into each anomaly with related logs, log properties, and AI-powered analysis to quickly identify the root cause.

Audit Logs

Track changes to pipeline configurations, user management (add/delete), API keys, projects, and metrics ingestion controls through detailed audit logs inside settings.

OpsAI Enhancements

Run Kubernetes commands directly from the investigation chat with a single click to resolve Kubernetes issues faster.

Create dashboards and alerts directly from the OpsAI chat by simply describing what you need.

S3 Routing

Route metrics, logs, and traces directly to your own AWS S3 bucket using the new self-service configuration page under Settings → S3 Routing.

We’re excited to introduce Enhanced LLM Observability in Middleware, with new capabilities that help you monitor, test, evaluate, and optimize your AI applications from a single platform.

What’s New

LLM Traces
Gain complete visibility into every LLM request with detailed traces, including tokens, latency, cost, error rates, stack traces, spans, and execution metadata to quickly troubleshoot issues.

Playground
Experiment with prompts across multiple LLM models, tools, and output schemas. Compare responses side by side and validate behavior before deploying to production.

LLM Evaluation
Measure response quality using configurable LLM-as-a-Judge evaluations. Define custom rules, choose the evaluation model, and automate response validation at scale.

GPU Monitoring
Monitor GPU hosts and devices with detailed metrics such as utilization, occupancy, memory usage, temperature, power consumption, bandwidth, and running processes.

LLM Dashboards
Track AI application performance with 50+ built-in dashboards covering token usage, latency, costs, LLM calls, evaluations, tool usage, and error rates.

Watch the Product Demo

Learn more about LLM Observability on our product page or get started in minutes with the setup guide.

Easy Integration

Monitor MongoDB deployments across self-hosted environments (Linux and Kubernetes) as well as MongoDB Atlas with a simple setup process.

Performance Monitoring

Track and monitor key database metrics, including connections, operation load, query throughput, latency, locks and concurrency, index performance, query execution, memory utilization, TCMalloc allocator metrics, and network activity.

Collection insights

Monitor collection-level metrics such as data size, storage size, index count, and object count to identify growth trends, optimize performance, and prevent storage-related issues.

Query metric analytics

Gain visibility into query statements, raw queries, cursor activity, locks, and flow control to identify slow queries, reduce contention, and improve overall database performance.

I’d love to hear your thoughts on this release. Simply reply to this email, and I’ll get back to you as soon as possible.

MCP Server Integration

• You can now use Middleware’s MCP Server to fetch observability data directly within your IDE, for faster troubleshooting.

• Ask questions about your observability data directly from your IDE and get errors and code context.

• Resolve application performance issues without leaving your IDE.

Refer to the integration documentation to get started.

• Easy Integration

  • Now supports self-hosted environments (Linux & Kubernetes), as well as Azure, GCP, and AWS.
    

• Performance Monitoring

  • Ensure database performance with key metrics such as throughput, execution duration, commits, rollbacks, connections, rows modified by operation type, and transaction time, blocks, vacuum operations.
    

• Query Metrics

  • Easily track low performing query with metrics such as execution counts, avg duration, shared blocks, related APM services , blocking activity etc.
    

• Schemas

  • View database schemas with definitions, columns, indexes, foreign keys along with schema level variants and metrics.
    

    
    

We would love to hear your thoughts on this release. Please write us at support@middleware.io

• Save time by automatically investigating issues across Kubernetes, APM, and RUM.

• Select the specific Kubernetes clusters, APM services, and RUM applications you want to auto investigate.

• Get notified for every auto-fixed issue through your preferred notification channel.

I would love to hear your thoughts on this release. Please write to us at support@middleware.io and I will get back to you as soon as possible.

OpsAI brings:

• End-to-end issue detection across your stack
• Automatic root cause analysis for faster troubleshooting
• Auto-fix capabilities to eliminate manual intervention

With OpsAI, teams can:

• Save up to 80% of engineering time

• Resolve issues with 90% accuracy

Want to dive deeper?
👉 Read the full PR announcement here
👉 Explore how it works in our blog.

If you’d like a quick demo or have questions, reach out at support@middleware.io or book a quick call on my calendar.

Topic: Achieving 360° Visibility in Azure with OpenTelemetry using Middleware
Speaker: Keval Bhogayata
Time: 5th May at 9:30 PM - 10:30 PM (UTC+05:30)

In this session, you’ll learn how to instrument Azure-based cloud-native workloads using OpenTelemetry across logs, metrics, traces, RUM, and Kubernetes—enabling complete, end-to-end observability for your applications.

👉 Register here : Microsoft Reactor
👉 Webinar details: Middleware

Looking forward to your participation!

Product Analytics with RUM

• You can now track user behavior, traffic, and engagement trends in one place.

• We’ve added Sankey charts to help you to visualize how users move through your platform.

• Added adoption and retention metrics so you can spot where users are growing and where they’re dropping off.
  Introducing Product Analytics: Unlocking User Insights! 📊 - Watch Video

  

Pipeline for Serverless

We’ve made it easier to work with logs in serverless environments. This includes

• JSON parsing for structured logs.

• Add custom labels for easier querying.

• Sensitive data masking for security.
  

Integration Enhancements

A few meaningful upgrades across the integrations you use most:

• Kubernetes Auto-Instrumentation — Now supports both OTel-native and eBPF modes, with runtime and executable-level configuration.

• Prometheus — Use pod regex to scrape metrics from specific databases (Redpanda, MongoDB, PostgreSQL) without pulling the full cluster.

• PostgreSQL — Added event and query sample collection, top query views for performance insights, and connection pool metrics in dashboards.

I would love to hear your thoughts on this release. Please write us at support@middleware.io and I will get back to you as soon as possible.

RUM Heatmap

• Analyze user behavior with heatmap.

• identify high-click areas and rage clicks.

• Correlate frontend activity with backend spans from Heatmap.

Integration Improvements

New dashboards for multiple GCP services:

• Cloud Run

• Load Balancing

• Pub/Sub

• Cloud Tasks

• Firestore

• Filestore
  
  

• Redis: Add multiple endpoints within a single machine during integration.

• JVM: JVM metrics dashboard with additional performance metrics.

• Kafka: Support to fetch cluster-wide broker metrics from a single node in multi-node Kafka clusters, eliminating the need for per-node integrations.

Alerts
Support to interpolation for host-based alerts to handle metric gaps:

• Linear – Fills gaps by drawing a straight line between nearby values.

• Last – Uses the last observed value to fill missing data.

• Default Zero – Replaces missing values with zero.

Improvements

• Saved log filters in logs now support search and delete.

• Pie charts in dashboards now display metric names on hover.

• Notebooks functionality has been moved under Dashboards.

I would love to hear your thoughts on this release. Please send us an email at support@middleware.io, and I will get back to you as soon as possible.

Middleware has been selected for the Microsoft for Startups Pegasus Program 2026 and is now live on the Azure Marketplace.

What this means if you’re on Azure

If you’re on Azure, you can now deploy Middleware directly through the Marketplace. Purchases count toward your Microsoft Azure Consumption Commitment (MACC), so you can use your existing Azure budget. Middleware will also appear as a single line item on your Microsoft invoice; no separate vendor onboarding needed.

This also means closer collaboration with Microsoft’s engineering and cloud teams, which translates to faster product improvements and deeper Azure integrations for you.

As always, Middleware provides your team with full root-cause analysis in under 2 minutes, resolves around 70% of infrastructure issues automatically, and uses flat data-ingestion pricing with unlimited users and hosts.

Read more about the Pegasus program.

Thanks for being part of the journey.

Create Dashboards with AI

- Build a dashboard within a few seconds. Blazing fast.
- No manual query or selecting from the component
- Completely free and unlimited usages

Read full documentation here.

Notebooks

- Quick sharable link to your stakeholder and team member.
- Correlate infrastructure metrics with business metrics.
- All the data is accessible in one place

Read full documentation here.

Endpoint Monitoring

- Monitor all external and internal endpoint at one place
- The best way to debug the error on the endpoint
- Easy to find latency

Read full documentation here.

Observability Pipeline

- Control data ingestion. Ingest what you need.
- Control the cost
- Filter and segregate the data within one place

Read documentation here.

Integration Updates – Vercel & Flutter RUM

• Vercel integration with Middleware.
  - One-click installation
  - Collect all the data like logs, metrics, and traces
  - No code changes required
  Read documentation here
  

• Flutter Integration for RUM
  - Now we support the Flutter SDK to collect all the rum sessions.
  - Session recording
  - Crash and error reporting
  Read documentation here.

Other improvement on platform

• 20 new dashboards of GCP services.

• 10 new default dashboards for AWS services

• Prometheus integration improvement.

• Linux auto instrumentation with Nodejs, .net, and Java (others coming soon!)

• Microsoft Entra ID for SSO support

• NOW SUPPORT AVAILABLE VIA INTERCOM LIVE CHAT

I would love to hear your thoughts on this release. Please send feedback on support@middleware.io

Vercel integration (upgraded)
You can now install Middleware on Vercel with one click — no code changes required. The app automatically configures log and trace drains so you can view correlated logs and traces out of the box.

We’ve also added support for the Redirect log source, in addition to Static, Lambda, Edge, Build, External, and Firewall.

To get started, check out the documentation here.

RUM for Flutter (new)
You can now integrate Middleware RUM into your Flutter apps to monitor real user performance. Get session recordings, performance metrics, and correlate front-end errors with backend traces to pinpoint issues and auto-fix them using Ops AI.

To configure the flutter application, check the documentation here.

If you’d like a walkthrough or help enabling this feature, just send us an email at support@middleware.io

You can define pipeline rules to manage how logs, metrics, and traces are processed and ingested into Middleware. This allows you to filter unnecessary data, apply ingestion controls, and keep only what matters.

The result: lower data volume, reduced costs, and better control over your observability footprint.

To get started, check out the documentation here.

If you’d like a walkthrough or help enabling this feature, just reply to this email.

You can monitor endpoints directly from your application or upload a list of endpoints manually. Middleware provides key performance metrics such as request rate, success rate, response time, dependencies, errors, etc.

This helps you proactively detect issues, understand endpoint behavior, and ensure your application runs smoothly with reliable API performance.

To get started, check out the documentation here.

If you’d like a walkthrough or help enabling this feature, just drop us an email at support@middleware.io

You can now create notebooks to combine logs, metrics, and traces, and add notes for full context. This makes it easy to analyze incidents and performance issues in a single, shareable document.

Notebooks support multiple inputs like text, tables, and data widgets, and can be quickly shared with your team for collaboration.

To get started, check out the documentation here.

If you’d like a walkthrough or help enabling this feature, send an email at support@middleware.io

With Middleware, you no longer need to manually select metrics, attributes, filters, or aggregations to build dashboards. Just write a prompt, and Middleware will generate the dashboard for you.

You can later edit, refine, or add more widgets using AI or the traditional manual approach.

Build dashboards faster, without SQL query and no guesswork.

To get started, check out the documentation here.

If you’d like a walkthrough or help enabling this feature, drop us an email at support@middleware.io

Global Search 🔍

Search the entire platform using Ctrl + K / Cmd + K to find:

• Screens (APM, traces, endpoint monitoring, dashboards, etc.)

• Actions (create alert, create dashboard, configure integrations, etc

Simplified Kubernetes Auto-Instrumentation ☸️

• Install with a single script — no extra configuration required.

• Middleware agent now automatically detects databases in the cluster, so you can configure monitoring directly.

  
  Installing Middleware Agent in a Kubernetes Cluster 🚀 - Watch Video

  

Linux Auto-Instrumentation 🐧

The agent can now:

• Detect running services on Linux hosts

• Automatically enable APM for Java, Node.js, and Python services

Streamlining Application Instrumentation on Linux Hosts 🚀 - Watch Video

Alert Templates 🚨

• Select a template and the alert is automatically configured with best-practice defaults.

• Templates are available for the Kubernetes, Infrastructure, APM, and Logs categories, with more categories coming soon.

  Creating Alerts with Templates for High CPU Usage 🚨 - Watch Video

  

​

APM Setup Improvements ⚙️

• Automatically suggests setup for alerts, logs, infra, and more based on detected services.

• See whether alerts, synthetic tests, logs, or metrics already exist for a service.

Dashboards 📊

• Added 15 new Kubernetes dashboards to the default set, including:

  • Ingresses, Services, PVs, PVCs

  • HPA, StatefulSets, DaemonSets

  • CronJobs, Jobs, ReplicaSets

  • Deployments, Containers, Namespaces, Clusters

RUM Updates 🌐

• Added support for cookieDomain in the SDK.
  Added new attributes:

  • Referrer domain

  • Page view

  • Page leave (on location change)

We’re launching a set of exciting new features next week! Stay tuned for our launch week starting February 2, 2026.

We’ve rolled out some important updates — from log monitoring to APM, access control, and security improvements.

Something huge is coming to Ops AI — it’s going to change observability forever. Book your beta session now and stay tuned!

Here’s what’s new 👇

🪵 Logs

• Surrounding Logs: View logs with the same attribute directly using the new “Surrounding Logs” option — perfect for faster troubleshooting.

  

• Pinned Attributes: Pin key attributes in expanded log view so they always stay on top.

  

• Custom Views: Choose columns from the attribute list and save them as “Views” for quick filtering later.

⚙️ APM

• Improved charts across all APM screens for better visibility.

• Automatic Java Profiling is now enabled out of the box.

• Python support added for Google Cloud Run.

• Added a new Service Metrics Dashboard tab for direct access from APM → Service.

  

🧑‍💼 Role-Based Access Control

• Admins can now assign project-level access to specific users.

• Set granular permissions for:

  • Creating or editing dashboards

  • Viewing installation scripts and API keys

• While inviting users, select or edit which projects they can access.

🐇 Middleware on AWS marketplace

• Middleware is now available on the AWS Marketplace →

• You can now sign up directly via AWS Marketplace and set up billing through your AWS account — no need to add payment details separately in Middleware.

🔐 Azure Intra ID SSO

• Login via SSO with Azure intra ID. See full documentation here.

• Admins can now add new users directly from the IDP — no need for manual invitations.

🛡️ Account Security Enhancements

• Inactive sessions (30+ days) now expire automatically.

• Any critical account change (like role or permission updates) will require re-login.

• Password resets will now log out all active sessions across devices.

🎨 UI/UX Updates

• Improved filters in the dashboard builder.

• Enhanced search and fixes on the RUM side.

• Refreshed Login and Signup screen design.

🌐 Browser Test Pricing

• Free plan: 10 browser test runs.

• Pay-as-you-go: $9.99 per 1K browser test runs.

• View detailed pricing on our pricing page.

We’re continuously working to make Middleware faster, smarter, and easier to use for developers and DevOps teams.

A major announcement is imminent! Over the next few weeks, we’re unveiling a breakthrough in Ops AI that will fundamentally transform observability. Prepare to see the future—our beta session schedule is opening soon. Stay tuned for booking details!

Have feedback or want to see a demo of the new features? Just drop an email at support@middleware.io— we’d love to hear from you.

Best,
The Middleware Team