Skip to content

Conversation

@androng
Copy link

@androng androng commented Mar 8, 2025

  • Add more Apify scrapers for Indeed, LinkedIn, Ziprecruiter
  • Make Apify scrapers run in parallel
  • Replace "Job summary" with "reasons for score"
  • Convert to OpenAI batch processing because there's a 50% discount and latency is acceptable
  • Convert to Typescript to lower the chance of runtime errors
  • Make README more clear
  • Make each Google Sheets API call add 1000 rows to avoid API rate limit
  • Put validation at the beginning of the program to avoid runtime errors
  • Did not think about Deepseek, may still work but not tested

…ove a bug "jobURL" in the deduplication that made jobs data never match
Change columns because I thought "job pay" and "job source" are more important than "posted date", add instructions for adding job preferences (removed the user preferences from the prompt), combined deepseek and openAI columns to save space, added the first "parser"
…to OpenAI at once to lower OpenAI bill, converted everything to Typescript so I could use Pqueue to run Apify actors in parallel
…each other, only deduplicated on jobs from spreadsheet. Fixed another bug where the new file format caused reading from file to fail
…t score" and the "employer fit score" to make the grading less ambiguous/more consistent
…idation for missing files, adding example files, make OpenAI client lazy loaded to make my validation come first. Also delete the job summary code because I didn't use it or maintain it.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant