# Closing the year Strong: December Product Updates

LangWatch has a completely new dashboard and visual design, Evaluations, traces, and signals are now front and center. the LangWatch team wants to wish you happy holidays and a healthy, successful 2026.

*By Manouk Draisma · December 24, 2025*

Canonical: https://langwatch.ai/blog/closing-the-year-strong-december-product-updates

As we wrap up the year, the **LangWatch** team wants to wish you happy holidays and a healthy, successful 2026.

Here’s to a year of agents that actually perform in production reliably, safely, and with confidence.

![Closing the year Strong: December Product Updates](https://framerusercontent.com/images/MvLcOs8J81p33PuB1hRfbsOIdE.png)

Over the past few months, we’ve increasingly heard the same thing from teams across industries: **LLMOps is becoming a core pillar of 2026 roadmaps**. If that’s true for your organization as well, feel free to reach out. Our team is happy to support you with input, examples, and structure for building a strong internal business case.

To close out the year, we shipped a set of meaningful product improvements. Not incremental tweaks, but changes that make LangWatch easier to use, easier to scale, and more powerful for teams running real-world agent systems.

Below is a quick overview of what’s new.

## A redesigned dashboard and UI

We’re ending the year with a big one: **LangWatch has a completely new dashboard and visual design**.

This isn’t a light refresh or cosmetic polish. We rethought how the product should feel when you’re actually using it day to day - especially when you’re debugging, evaluating, or investigating issues in production.

Evaluations, traces, and signals are now front and center, with:

-   Clearer structure and hierarchy
    
-   Better information density without overwhelming the screen
    
-   Far less visual friction when analyzing results
    

The goal was simple: reduce the time between *something looks off* and *I understand exactly why*.

If you haven’t logged in recently, this is a great moment to take another look.

![LLM evaluations dashboard](https://framerusercontent.com/images/iF7BGKPYIaM8XPEjCX6jdBb4Pc.png)

## Replicate workflows, prompts, evaluations, and datasets across projects

If you’re running multiple LangWatch projects - for example, separate environments for development, staging, and production this update was built specifically for you.

You can now **replicate workflows, prompts, evaluations, and datasets across projects**, giving you far more control over how you manage real-world setups:

-   Maintain full isolation and access control between dev, staging, and prod
    
-   Reuse the *exact same* evaluation logic across environments
    
-   Replicate only what you need, when you need it
    
-   Iterate and sync without duplicating effort or losing consistency
    

This makes LangWatch significantly easier to operate at scale, especially for teams running multiple agents or pipelines in parallel.

![Closing the year Strong: December Product Updates](https://framerusercontent.com/images/oqpNYZIBF7BKRuUW81tPMGwqItw.png)

## Analytics updates for more advanced monitoring

Analytics has also received a series of improvements aimed at more advanced organization and monitoring workflows.

You can now:

-   **Create multiple dashboard pages per project**  
    Separate analytics by use case, agent, evaluation, or environment.
    
-   **Reorder and manage dashboard pages**  
    Keep the most relevant insights easily accessible.
    
-   **Group traces by labels within an evaluation**  
    Enable consistent metric comparisons across categories or scenarios.
    
-   **Attach alerts to analytics graphs**  
    Trigger notifications when metrics cross thresholds or regress against a baseline.
    

Together, these updates make it easier to move from passive observability to proactive monitoring - and to catch issues before they impact users.

## Upcoming webinar: testing agentic AI the right way

On **13 January**, we’re hosting a technical webinar titled:

[**“LLM evals aren’t enough: Testing agentic AI the right way.”**](https://luma.com/vevgrewo)

Rogerio will be joined by **Ron Kremer (PhD AI, ADC Data & AI)** for a deep, practical discussion on how enterprise teams should think about evaluation and testing in the age of agentic systems.

If you’re building or deploying agents beyond simple single-step prompts, this one’s worth attending.

## Latest content from LangWatch

If you missed some of our recent content, here are a few highlights to close out the year:

-   [A recap of our most recent Launch Week](https://langwatch.ai/launch-week-nov)
    
-   [*From 0 to 1.4k GitHub stars in a weekend*: the Better Agents Manifesto](https://langwatch.ai/better-agents-manifesto)
    
-   [*Why agentic AI needs a new layer of testing*](https://langwatch.ai/blog/why-agentic-ai-needs-a-new-layer-of-testing)
    

More deep dives, hands-on guides, and agent testing content are coming early in 2026.

Thanks for building with LangWatch this year. We’re excited about what’s ahead - and we’re looking forward to helping you ship **better, safer, and more reliable agents** in the year to come.

Happy holidays,  
**The LangWatch Team**