[HN Gopher] Show HN: JobLens - AI-powered job search for 'Who Is...
       ___________________________________________________________________
        
       Show HN: JobLens - AI-powered job search for 'Who Is Hiring'
        
       There are existing HN job aggregators, but I thought we could take
       it a step further. Inspired by an insightful comment on a previous
       thread (https://news.ycombinator.com/item?id=36163021), I built a
       tool that aggregates job postings and intelligently categorizes
       them based on user-specific preferences:  * Country and remote work
       preferences  * Employer type (e.g., startup, corporation,
       government)  * Industry  * Technologies used  * Role type
       (developer, architect, product owner, etc.)  * Salary range (where
       available)  One of the superpowers of LLMs is reformatting
       information from any format X to any other format Y. We leverage
       this to map all the unstructured job postings into the same unified
       structure. The new GPT functions feature and the extended context
       windows are really helpful for this. Instead of having to build a
       custom NER pipeline, it works very well with GPT out-of-the box.
       One challenge is keeping the filters consistent and merging of
       duplicates. Embeddings help with that.  What's next:  * Integrate
       additional sources. We can generate web scrapers and data
       processing steps on the fly that extract and transform data into
       the same structure.  * Add location distance filters.  * Expand
       beyond jobs to monitor personalized data like events or real
       estate. Imagine using AI to rate local events from multiple sources
       based on your preferences, considering factors like your interests
       and distance from home.  * Smaller improvements based on your
       feedback :)
        
       Author : hubraumhugo
       Score  : 46 points
       Date   : 2023-07-03 15:16 UTC (7 hours ago)
        
 (HTM) web link (www.kadoa.com)
 (TXT) w3m dump (www.kadoa.com)
        
       | SCUSKU wrote:
       | Wow this is amazing. This is more or less what I was trying to
       | build https://hnresumetojobs.com to look like, I am reconsidering
       | whether or not to continue now.
       | 
       | I tried using the OpenAI functions API to do structured text
       | extraction, but found that it would hallunicate a lot of things.
       | Do you guys have that problem? How did you guys go about solving
       | that?
        
       | baxtr wrote:
       | Nice! Is there something similar for freelancing gigs?
        
       | boxcarr wrote:
       | JobLens is cool! This past month we had the same idea -- it's
       | been a fun project: https://hnjobs.u-turn.dev
       | 
       | ChatGPT does an incredible job parsing, but then lots of effort
       | goes into normalizing and deduping each field. Long story short,
       | your results look quite good to me!
        
         | hubraumhugo wrote:
         | Your project looks quite impressive as well, especially the
         | extracted URLs to apply and the candidate profiles, didn't get
         | that far yet :) Automating tedious work like data extraction
         | and transformation is a great use case for LLMs.
        
           | boxcarr wrote:
           | Not sure if you found this as well, but gpt-3.5-turbo-016
           | does a poor job following instructions other than parsing.
           | So, to work around this, we prompt gpt-3.5-turbo with the
           | rules we want applied say an extracted field and then go back
           | to gpt-3.5-turbo-016 to parse with chatgpt functions.
           | 
           | Bottom line, every single post requires approximately 10
           | different prompts to refine the extraction.
        
       | rahimnathwani wrote:
       | This is cool. I'm curious about this part:                 We
       | leverage this to map all the unstructured job postings into the
       | same unified structure. The new GPT functions feature and the
       | extended context windows are really helpful for this.
       | 
       | Does this mean that, rather than using the gpt-3.5-turbo chat
       | API, and using a system prompt that describes the output format
       | you want (JSON?), you instead describe your output format in the
       | same way you would if you were going to call a function?
        
         | hubraumhugo wrote:
         | Yes, you can essentially use GPT functions to reliably output
         | JSON based on a predefined schema. See
         | https://github.com/openai/openai-cookbook/blob/main/examples...
        
       | timhigins wrote:
       | Site is down for me, maybe the HN hug of death? Anyone else
       | seeing this?
        
       ___________________________________________________________________
       (page generated 2023-07-03 23:01 UTC)