Build AI agents by describing your task in plain English. Vellum turns it into a working agent you can run on a schedule, trigger via API, or use through a UI. No code, no setup; just automation you understand and control.
What stood out to me about Vellum is how intentionally it frames complexity instead of trying to hide or over-simplify it.
In tools that deal with structured thinking or layered workflows, it’s common to either overwhelm users or abstract things too much. Vellum feels like it’s aiming for a middle ground — making the structure visible, but not intimidating. From a UX perspective, that balance usually comes from deeply understanding where users get confused versus where they actually want control.
As a first impression, the product feels considered and mature, prioritizing clarity over flash, which makes it easier to trust and grow into over time.
Report
This is exactly what the AI agent space needs! The ability to describe workflows in plain English and have them turn into production-ready agents is a game changer. I'm particularly impressed by the state management approach mentioned by @dvargas92495 - that's crucial for building reliable agents that can handle complex multi-step workflows.
For teams building LLM applications, having the transparency to preview each step and review past runs addresses one of the biggest pain points: trust and debuggability. Too many agent frameworks are black boxes.
Quick question: How does Vellum handle API rate limiting and retries when agents are calling external services? This is often a challenge in production agent deployments.
Report
This looks incredibly interesting. I love the landing page design and the example workflow for the SEO blog post was particularly compelling. I was recently working on SEO implementation for a website and this seems like exactly what I need. I'm excited to explore the other automation possibilities available!
I just built an agent that scrapes Twitter for trending topics and sends me daily post ideas and it took me like 10ish min. This is seriously so cool. Huge congrats on the launch, team!! The new agent builder is 🔥
This is such an important unlock. Agents have been incredibly powerful for engineers, but the rest of the org has been stuck duct-taping prompts together. Making agents usable for non-technical teams is where real adoption happens. Curious how Vellum handles edge cases and process changes over time without breaking everything?
Minara
What stood out to me about Vellum is how intentionally it frames complexity instead of trying to hide or over-simplify it.
In tools that deal with structured thinking or layered workflows, it’s common to either overwhelm users or abstract things too much. Vellum feels like it’s aiming for a middle ground — making the structure visible, but not intimidating. From a UX perspective, that balance usually comes from deeply understanding where users get confused versus where they actually want control.
As a first impression, the product feels considered and mature, prioritizing clarity over flash, which makes it easier to trust and grow into over time.
This is exactly what the AI agent space needs! The ability to describe workflows in plain English and have them turn into production-ready agents is a game changer. I'm particularly impressed by the state management approach mentioned by @dvargas92495 - that's crucial for building reliable agents that can handle complex multi-step workflows.
For teams building LLM applications, having the transparency to preview each step and review past runs addresses one of the biggest pain points: trust and debuggability. Too many agent frameworks are black boxes.
Quick question: How does Vellum handle API rate limiting and retries when agents are calling external services? This is often a challenge in production agent deployments.
This looks incredibly interesting. I love the landing page design and the example workflow for the SEO blog post was particularly compelling. I was recently working on SEO implementation for a website and this seems like exactly what I need. I'm excited to explore the other automation possibilities available!
The 2025 state of AI Development
@rohanrecommends - that's great to hear! we can give you a sample SEO workflow later today that you can clone to your workspace and get started
Huge congrats on the launch — love how Vellum turns plain-English workflows into safe, production-ready AI agents for non-technical teams.
The 2025 state of AI Development
@zeiki_yu thanks Zeiki!! Excited to launch it
The 2025 state of AI Development
@zeiki_yu - that's exactly right! give it a shot and let me know what you think
Congrats to the whole team 👏👏
The 2025 state of AI Development
I just built an agent that scrapes Twitter for trending topics and sends me daily post ideas and it took me like 10ish min.
This is seriously so cool. Huge congrats on the launch, team!! The new agent builder is 🔥
The 2025 state of AI Development
@marinatrajkovska share the agent with us - we can share it with the community!
Respell
This is such an important unlock. Agents have been incredibly powerful for engineers, but the rest of the org has been stuck duct-taping prompts together. Making agents usable for non-technical teams is where real adoption happens. Curious how Vellum handles edge cases and process changes over time without breaking everything?