Vellum is coming to the AI Engineering World's Fair in SF. Come visit our booth and get a live demo!

Vellum Product Update | September

Agent Builder (beta), Custom Nodes, AI Apps, and more for faster and more complex agent building in Vellum.

7
Written by
Reviewed by
No items found.

The only thing scarier than spooky season this October is the amount of updates we shipped to Vellum in September.

This month’s updates make agent building faster and easier than ever with the beta launch of Agent Builder, AI App UI for workflows, and Custom Node creation directly in the UI.

We’ve got lots more features and updates packed in, so let’s go straight in!

Agent Builder (beta)

Building an agent used to mean dragging and dropping nodes wiring prompts, tools, and UI by hand, then iterating through lots of trial-and-error.

Now spinning up any agent is as easy as prompting Vellum’s Agent Builder with your task and use case. Add in all your context and necessary components to build your agent in minutes.

Agent Builder can:

  • Build agents
  • Debug workflows
  • Optimize prompts

All with interactive components surfaced directly in chat to let you trigger workflow and populate suggestions without leaving the chat.

This beta feature is available in all Workflow sandboxes for you to try today! Read more about Agent Builder here.

AI App UI for Deployed Workflows

Previously, the only way to use a Workflow once deployed was to call it via API. This meant you'd have to loop in engineers to build a whole backend and frontend to interact with your Workflow.

Now every deployed Workflow can be automatically turned into AI App – complete with a UI that allows you to run and test Workflows in the browser without writing code.

Vellum analyzes your deployed Workflow to generate the AI App UI with the appropriate input fields for plug and play use.

This makes it easier than every to spin up AI-powered internal tools and automations, as well as prototypes of new AI-powered user experiences. Learn more in our announcement!

Create Custom Nodes from the UI

Before, turning bespoke logic into a reusable nodes meant pushing up the code representation of a Workflow.

Custom Nodes can now be created and customized directly in the Workflow UI. With the node’s Open Definition button, you can preview and edit it’s code without leaving Vellum.

Available in the side panel, Custom Nodes allow you to define any custom behavior by specifying:

  • Input attributes
  • Outputs to return
  • Underlying code to execute

This makes it easier to turn one-off logic into reusable building blocks, speeding up iteration and keeping workflows cleaner. Learn more here.

Workflows & Prompts: Image & Document Inputs

Passing documents and images into your workflows used to be a meticulous process of using the (admittedly unintuitive) "Chat History" variables.

Now, images and documents can be passed directly as first-class input variables in both Workflows and Prompts.

With this, it's much easier to build out use cases for data extraction from pdfs, image analysis, and more.

Workflows

Agent Node Code Tool Custom Inputs

Before, the code tool could only received what the model populated, limiting control.

You can now pass Custom Inputs to the Agent Node’s Code Tool, even if they aren’t populated by the model.

This gives you deterministic control over tool execution, so you can inject IDs, configs, or other critical parameters without depending on model output.

Deployments

Stack Trace for Workflow Execution Errors

Error surfaces in Vellum used to be shallow, only showed high-level messages with little detail.

Now, you can view the full stack trace for Workflow execution errors directly in Vellum.

This shortens mean time to resolution by making it easy to pinpoint exactly where things went wrong.

Raw Data for Provider Errors

Before, failures were opaque and you couldn’t see the provider’s raw payload after executions.

Now, you can view the provider’s raw response whenever a prompt fails.

This makes it much easier to diagnose issues like rate limits, schema mismatches, or safety filters, as well as drive downstream logic based on this data.

New Model Support

  • Google
    • Gemini 2.5 Flash
    • Gemini 2.5 Flash Lite
  • OpenAI
    • GPT‑5 Codex Model

Quality of Life Improvements

SDK Preview Code Search

It used to be tedious to find the right method or type by scanning the generated SDK preview files.

Now, you can search across filenames and file contents with highlighted matches, making code integrations faster and smoother.

Custom Node Colors & Icons

Before, there was no way to differentiate custom nodes in your Workflows.

Now, you can customize each node with colors and icons to visually group logic, making scanning and code reviews within Workflows easier.

Markdown Editor in Note Nodes

Notes in workflows used to be restricted to plain text, making documentation unintuitive and hard to read.

Note Nodes now fully support full markdown, including headers, code blocks, iframes, and more. Markdown enabled notes let you place context directly next to your workflow logic, to make documentation more practical and thorough.

That’s a wrap for September

That’s a wrap for September. From agent building to streamlining custom code integrations, this round of updates is all about making building easier and faster for the whole team to use Vellum.

We’ll be back next month with more improvements, excited to hear your feedback on these!

{{general-cta}}

ABOUT THE AUTHOR
Noa Flaherty
Co-founder & CTO

Noa Flaherty, CTO and co-founder at Vellum (YC W23) is helping developers to develop, deploy and evaluate LLM-powered apps. His diverse background in mechanical and software engineering, as well as marketing and business operations gives him the technical know-how and business acumen needed to bring value to nearly any aspect of startup life. Prior to founding Vellum, Noa completed his undergrad at MIT and worked at three tech startups, including roles in MLOps at DataRobot and Product Engineering at Dover.

ABOUT THE reviewer
Nicolas Zeeb
Technical Content Lead

Nick is Vellum’s technical content lead, writing about practical ways to use both voice and text-based agents at work. He has hands-on experience automating repetitive workflows so teams can focus on higher-value work.

No items found.
lAST UPDATED
Oct 1, 2025
share post
Expert verified
Related Posts
Guides
September 30, 2025
15
A practical guide to AI automation
LLM basics
September 25, 2025
8 min
Top Low-code AI Agent Platforms for Product Managers
LLM basics
September 25, 2025
8 min
The Best AI Agent Frameworks For Developers
Product Updates
September 24, 2025
7 min
Introducing AI Apps: A new interface to interact with AI workflows
LLM basics
September 18, 2025
7 min
Top 11 low‑code AI workflow automation tools
All
September 16, 2025
12 min
MCP UI & The Future of Agentic Commerce
The Best AI Tips — Direct To Your Inbox

Latest AI news, tips, and techniques

Specific tips for Your AI use cases

No spam

Oops! Something went wrong while submitting the form.

Each issue is packed with valuable resources, tools, and insights that help us stay ahead in AI development. We've discovered strategies and frameworks that boosted our efficiency by 30%, making it a must-read for anyone in the field.

Marina Trajkovska
Head of Engineering

This is just a great newsletter. The content is so helpful, even when I’m busy I read them.

Jeremy Hicks
Solutions Architect

Experiment, Evaluate, Deploy, Repeat.

AI development doesn’t end once you've defined your system. Learn how Vellum helps you manage the entire AI development lifecycle.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Build AI agents in minutes with Vellum
Build agents that take on the busywork and free up hundreds of hours. No coding needed, just start creating.

General CTA component, Use {{general-cta}}

Build AI agents in minutes with Vellum
Build agents that take on the busywork and free up hundreds of hours. No coding needed, just start creating.

General CTA component  [For enterprise], Use {{general-cta-enterprise}}

The best AI agent platform for enterprises
Production-grade rigor in one platform: prompt builder, agent sandbox, and built-in evals and monitoring so your whole org can go AI native.

[Dynamic] Ebook CTA component using the Ebook CMS filtered by name of ebook.
Use {{ebook-cta}} and add a Ebook reference in the article

Thank you!
Your submission has been received!
Oops! Something went wrong while submitting the form.
Button Text

LLM leaderboard CTA component. Use {{llm-cta}}

Check our LLM leaderboard
Compare all open-source and proprietary model across different tasks like coding, math, reasoning and others.

Case study CTA component (ROI)

40% cost reduction on AI investment
Learn how Drata’s team uses Vellum and moves fast with AI initiatives, without sacrificing accuracy and security.

Case study CTA component (cutting eng overhead) = {{coursemojo-cta}}

6+ months on engineering time saved
Learn how CourseMojo uses Vellum to enable their domain experts to collaborate on AI initiatives, reaching 10x of business growth without expanding the engineering team.

Case study CTA component (Time to value) = {{time-cta}}

100x faster time to deployment for AI agents
See how RelyHealth uses Vellum to deliver hundreds of custom healthcare agents with the speed customers expect and the reliability healthcare demands.

[Dynamic] Guide CTA component using Blog Post CMS, filtering on Guides’ names

100x faster time to deployment for AI agents
See how RelyHealth uses Vellum to deliver hundreds of custom healthcare agents with the speed customers expect and the reliability healthcare demands.
New CTA
Sorts the trigger and email categories

Dynamic template box for healthcare, Use {{healthcare}}

Start with some of these healthcare examples

SOAP Note Generation Agent
Personalized healthcare explanations of a patient-doctor match

Dynamic template box for insurance, Use {{insurance}}

Start with some of these insurance examples

AI agent for claims review and error detection
Insurance claims automation agent
Collect and analyze claim information, assess risk and verify policy details.

Dynamic template box for eCommerce, Use {{ecommerce}}

Start with some of these eCommerce examples

E-commerce shopping agent

Dynamic template box for Marketing, Use {{marketing}}

Start with some of these marketing examples

Competitor research agent
Scrape relevant case studies from competitors and extract ICP details.

Dynamic template box for Legal, Use {{legal}}

Start with some of these legal examples

PDF Data Extraction to CSV
Extract unstructured data (PDF) into a structured format (CSV).

Dynamic template box for Supply Chain/Logistics, Use {{supply}}

Start with some of these supply chain examples

Risk assessment agent for supply chain operations

Dynamic template box for Edtech, Use {{edtech}}

Start with some of these edtech examples

Turn LinkedIn Posts into Articles and Push to Notion
Convert your best Linkedin posts into long form content.

Dynamic template box for Compliance, Use {{compliance}}

Start with some of these compliance examples

No items found.

Dynamic template box for Customer Support, Use {{customer}}

Start with some of these customer support examples

Trust Center RAG Chatbot
Read from a vector database, and instantly answer questions about your security policies.

Template box, 2 random templates, Use {{templates}}

Start with some of these agents

Turn LinkedIn Posts into Articles and Push to Notion
Convert your best Linkedin posts into long form content.
Agent that summarizes lengthy reports (PDF -> Summary)
Summarize all kinds of PDFs into easily digestible summaries.

Template box, 6 random templates, Use {{templates-plus}}

Build AI agents in minutes

SOAP Note Generation Agent
Synthetic Dataset Generator
Generate a synthetic dataset for testing your AI engineered logic.
Personalized healthcare explanations of a patient-doctor match
Competitor research agent
Scrape relevant case studies from competitors and extract ICP details.
Trust Center RAG Chatbot
Read from a vector database, and instantly answer questions about your security policies.
Automated Code Review Comment Generator for GitHub PRs

Build AI agents in minutes for

{{industry_name}}

Competitor research agent
Scrape relevant case studies from competitors and extract ICP details.
AI agent for claims review and error detection
E-commerce shopping agent
Retail pricing optimizer agent
Analyze product data and market conditions and recommend pricing strategies.
Risk assessment agent for supply chain operations
Insurance claims automation agent
Collect and analyze claim information, assess risk and verify policy details.

Case study results overview (usually added at top of case study)

What we did:

1-click

This is some text inside of a div block.

28,000+

Separate vector databases managed per tenant.

100+

Real-world eval tests run before every release.