We've been using the platform for running our agents. We use the vercel ai sdk and electric sql. Been very happy with everything.
Dallin's building https://www.buster.so (agents for data analytics).
Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
41–50 of 70 posts
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#42The platform looks great! However I do personally really dislike that everyone is either marketing themselves or has truly pivoted to AI agents… This seems like a great platform to run any type of tasks.
> However I do personally really dislike that everyone is either marketing themselves or has truly pivoted to AI agents…
I'm the CEO of a direct competitor to trigger, and we're in the same boat. What we make is great for any type of workflow, but our investors are pushing us hard to frame it as AI agents.
FWIW, more than 1/2 of our customers are AI agents, so it does make sense. But you're right, it's great for all sorts of use cases (trigger and us).
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#43The platform looks great! However I do personally really dislike that everyone is either marketing themselves or has truly pivoted to AI agents… This seems like a great platform to run any type of tasks.
> However I do personally really dislike that everyone is either marketing themselves or has truly pivoted to AI agents… I'm the CEO of a direct competitor to trigger, and we're in the same boat. What we make is great for any type of workflow, but our investors are pushing us hard to frame it as AI agents. FWIW, more than 1/2 of our customers are AI agents, so it does make sense. But you're right, it's great for all…
"our investors are pushing us hard to frame it as AI agents"
One of the biggest downsides of raising money. The moment you have investors, they own a lot of what you do even though they shouldn't in theory.
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#44Earlier quoted context omitted.
> However I do personally really dislike that everyone is either marketing themselves or has truly pivoted to AI agents… I'm the CEO of a direct competitor to trigger, and we're in the same boat. What we make is great for any type of workflow, but our investors are pushing us hard to frame it as AI agents. FWIW, more than 1/2 of our customers are AI agents, so it does make sense. But you're right, it's great for all…
"our investors are pushing us hard to frame it as AI agents" One of the biggest downsides of raising money. The moment you have investors, they own a lot of what you do even though they shouldn't in theory.
In their defense, they are trying to help us grow and get follow on investment. I'm apprecative of my investor's input. They aren't forcing me to do anything, just highly encouraging.
All of the un-budgeted money right now is for AI related workloads, so if you want enterprise money you have to pitch your product as AI.
And if you want investor money right now, you better have an AI product.
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#45The platform looks great! However I do personally really dislike that everyone is either marketing themselves or has truly pivoted to AI agents… This seems like a great platform to run any type of tasks.
yeah this really confused me... at first I was like... oh boy yet another "agentic ai saas" but then looking closer I realized this is like hosted temporal alternative which made me a lot more curious.
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#46The platform looks great! However I do personally really dislike that everyone is either marketing themselves or has truly pivoted to AI agents… This seems like a great platform to run any type of tasks.
I do some AI agent work; I suspect it's because there's a lot of people with infrastructure and programming skills that are new to building "agents". It's easy to think this requires monolithic architecture when a distributed DAG is often optimal. A lot of frameworks and SDKs for "agents" also inadvertently encourage monolithic design.
Long way of saying it's the marketing machine doing its thing. This looks like Cloud or run it yourself DAG software with some bells and whistles.
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#47Quite happy when I got to play with it – but I do prefer Inngest as the code is hosted where our actual app is hosted; not on their server. So it's just easier (but it's a tradeoff). Good tool, good tooling, congrats to the team!
That's fair, we understand there can be trade-offs when picking a workflow tool. Was there anything else we were missing or was it purely the hosting model which made the difference?
We appreciate the support either way!
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#48I like Trigger but it's a lot of complexity... > use cases like compute-heavy tasks such as generating videos using AI (Icon.com), real-time computer use (Scrapybara), AI enrichment pipelines (Pallet, Centralize), and vibe coding tools (Hero UI, Magic Patterns, Capy.ai) Okay, but aren't these websites using Trigger to schedule remarketing slop? Like adding you to Slack, sending you an email on day 1, sending you an e…
co-founder of Magic Patterns here, saw were were mentioned, figured I'd chime in: We don't use Trigger for marketing at all and I actually never thought of it for that use case. We're an AI design tool - prompt to create an interactive mockup - and we use Trigger to take screenshots of designs to provide a preview image. Taking a screenshot sounds easy, but it's not because Puppeteer constantly hits OOM errors. So yo…
Can you expand a bit on this use case? Why would Puppeteer constantly crash on your own high end machine, but not on Trigger's infra? Puppeteer doesn't care where it runs, so it would be nice to understand how Trigger's infra works around this problem.
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#49Congrats on the launch guys. It's been inspiring to see you iterate on the original idea and get more capable in each version. Can you say more about "we found that most developers struggled to write reliable code with implicit determinism". What were some of the common mistakes you were seeing?
Very common issues were: forgetting to put non-determistic code inside of steps (deterministic code can be put outside of steps, but non-deterministic = boom), incorrect use of cache keys (people would put dynamic data inside the cache key). Another issue we hit pretty frequently that a single step would take longer than the serverless function timeout it was running on (this was before we ran our own compute platfor…
You keep saying deterministic… What are you actually trying to say?
Based on your docs: I mean googling “deterministic temporal.dev” brings up nothing. I found other libraries and I get it. I assume you mean, “replayable by our engine,” but that would give away: replayability of most code in all the languages you support is hopelessly out of scope. “Pure” isn’t even a useful constraint - lots of code looks pure, but basically none of Python is, because you will alloc by so much as sneezing and hence you can OOM. Why promise the world?
Re: Launch HN: Trigger.dev (YC W23) – Open-source platform to build reliable AI apps
#50The reliability piece is interesting - we've seen AI apps fail in ways that are hard to predict or reproduce. Traditional monitoring doesn't really work when your 'bugs' are stochastic.
How are you handling failure modes that only show up statistically? Like when a model starts giving subtly wrong answers 2% of the time after a deployment.