Is ai web research node limited to # of loops? (ex. 600s for ~10-25 searches?)

[michaelraspuzzi]

hi! i’m hoping to run the ai web research node on 500+ rows of data to go from:

1. input name + location to
2. output linkedin url (if it can find it of course).

when testing it out with ~5 searches this workflow worked.

when scaling up to 45 searches, it timed out and didn’t output anything. it said it timed out at 600s.

and so if 4 searches took ~90 seconds, does that mean ai web research will time out at around 25 searches at a time?

here’s photo of the error.

here’s the workflow: https://www.gumloop.com/pipeline?workbook_id=gmKWgoEyhX9Sc957mxTkYE&run_id=guRBT3fpffDGBPoRd4A3ia

[Gumloop_Bot]

Hey @michaelraspuzzi! If you’re reporting an issue with a flow or an error in a run, please include the run link and make sure it’s shareable so we can take a look.

1. Find your run link on the history page. Format: https://www.gumloop.com/pipeline?run_id={your_run_id}&workbook_id={workbook_id}

2. Make it shareable by clicking “Share” in the top-left corner of the flow screen.
   

3. Provide details about the issue—more context helps us troubleshoot faster.

You can find your run history here: https://www.gumloop.com/history

[Wasay-Gumloop]

Hey @michaelraspuzzi - If the AI Web Research node is not able to find any result it may timeout. You can use an Error Shield here to skip one-off failed runs without halting the entire flow.

Error Shield Tutorial: https://vimeo.com/1059013147

Alternatively, you can try the normal Web Search node as well and format your input query like this to find the Linkedin URL. You can use the Combine Text node to format your query:

{name} {location} site:linkedin.com/in

Let me know if these options work for you.

[michaelraspuzzi]

thank you for sending! experimenting with the error shield now and that does help.

also, i skipped the web search as was exploring sonar from perplexity and ai web research node. thank you for the tip.

the web search node is faster, and just as accurate at first eval.

super helpful, thanks again.

[Wasay-Gumloop]

No problem!