Sim is an open-source workspace to build agentic workflows. Connect your AI agents and workflows to 1,000+ integrations and LLMs.
Replies
Best
Congratulations Team Sim,
That 90% chat-over-canvas shift is fascinating 😅 We're in the automation space too, so I'm curious, can you explain a bit how Sim decides when to use a deterministic step vs an actual LLM call? That "don't burn tokens on work that didn't need an LLM" idea sounds simple but feels hard to get right in practice.
@mittalpatel thank you! The two types of nodes in the workflow are either deterministic "function", "API" or LLM-based "agent", "router", "evaluator", etc. Users can choose between either!
@kelly_king3 I completely agree! I hope you enjoy!
Report
@emirkarabeg The per-node flexibility between chat, canvas, and code is what other tools get wrong. Curious, which mode do most of your users end up living in?
I've seen everything from horse stables health tracking to home assistants to orchestrating thousands of manufacturing facilities. Honestly, you can do whatever you like.
Report
Congratulations on the launch!
the chat to agent workflow looks incredibly intuitive. how does Sim decide when to use deterministic steps versus an LLM during execution?
@avery_thompson2 it's entirely within your control to set up your workflow deterministically or LLM-based. If you'd like an agent to decide on the path, you can add a router that routes the workflow one way or another depending on the context you have at that point in time in the workflow.
Report
congrats! building agents through natural language feels like the right direction. how much manual editing is typically needed after Sim generates an agent?
@imogen_wallace if you come in with a good prompt, most workflows and agents are one-shottable!
Report
The token cost angle is underrated honestly 🚀 Everyone talks about agent capabilities but nobody talks about how expensive they get at scale. How much are teams typically saving once they switch to deterministic steps where possible?
@amna9 deterministic steps cut out the expensive tool call and processing steps that normally happen first when running an agent. Now, they only receive the context they need to produce a relevant output.
Report
the launch story really resonates because many builders have experienced tool sprawl. what was the hardest engineering challenge in bringing workflows, memory and integrations into one platform?
@olivia_bennett7 the most difficult part was scaling our workflow loads to be able to handle multi-hour or multi-day runs. Now, you can run workflows for days on end!
Report
Congrats on the launch! The shared workspace concept makes a lot of sense. can agents securely share context while maintaining role based permissions?
@barnaby_lloyd great question! You can create permission groups that lock down agent capabilities and give scoped access.
Report
replacing expensive LLM calls with deterministic logic is a smart design choice. do users get visibility into where tokens are being spent inside each workflow?
Replies
Congratulations Team Sim,
That 90% chat-over-canvas shift is fascinating 😅 We're in the automation space too, so I'm curious, can you explain a bit how Sim decides when to use a deterministic step vs an actual LLM call? That "don't burn tokens on work that didn't need an LLM" idea sounds simple but feels hard to get right in practice.
@mittalpatel thank you! The two types of nodes in the workflow are either deterministic "function", "API" or LLM-based "agent", "router", "evaluator", etc. Users can choose between either!
This is so needed - agents working in isolation just aren't as effective!
@kelly_king3 I completely agree! I hope you enjoy!
@emirkarabeg The per-node flexibility between chat, canvas, and code is what other tools get wrong. Curious, which mode do most of your users end up living in?
@maurya_abhiranjan most users use the chat to workflow feature. It's very powerful! Give it a try.
I'd love to see more examples of workflows that have been built with Sim. What kind of use cases are you envisioning for the 1,000+ integrations?
@aymnart
I've seen everything from horse stables health tracking to home assistants to orchestrating thousands of manufacturing facilities. Honestly, you can do whatever you like.
Congratulations on the launch!
the chat to agent workflow looks incredibly intuitive. how does Sim decide when to use deterministic steps versus an LLM during execution?
@avery_thompson2 it's entirely within your control to set up your workflow deterministically or LLM-based. If you'd like an agent to decide on the path, you can add a router that routes the workflow one way or another depending on the context you have at that point in time in the workflow.
congrats! building agents through natural language feels like the right direction. how much manual editing is typically needed after Sim generates an agent?
@imogen_wallace if you come in with a good prompt, most workflows and agents are one-shottable!
The token cost angle is underrated honestly 🚀
Everyone talks about agent capabilities but nobody talks about how expensive they get at scale. How much are teams typically saving once they switch to deterministic steps where possible?
@amna9 deterministic steps cut out the expensive tool call and processing steps that normally happen first when running an agent. Now, they only receive the context they need to produce a relevant output.
the launch story really resonates because many builders have experienced tool sprawl. what was the hardest engineering challenge in bringing workflows, memory and integrations into one platform?
@olivia_bennett7 the most difficult part was scaling our workflow loads to be able to handle multi-hour or multi-day runs. Now, you can run workflows for days on end!
Congrats on the launch!
The shared workspace concept makes a lot of sense. can agents securely share context while maintaining role based permissions?
@barnaby_lloyd great question! You can create permission groups that lock down agent capabilities and give scoped access.
replacing expensive LLM calls with deterministic logic is a smart design choice. do users get visibility into where tokens are being spent inside each workflow?
@emily_carter18 inside your logs, you're able to see every trace span and token cost.