Find jobs on Indeed with Claude, then track them in Lamr

What each connector does

At the time of writing, Indeed offers a connector for Claude, in beta. Once you add it and sign in with your Indeed account, Claude can search postings by title, keywords, location and job type, open the details of one posting, read your Indeed profile, and look up an employer's ratings and pay data. It gives you links to apply. It does not keep a list of what you looked at, and it does not track applications.

Lamr's connector is the other half. It holds your master resume, your profile and every application you have saved, and it can score your fit for a posting, draft a resume and cover letter from what is true about you, and compile them as PDFs.

Add both to the same Claude account and one conversation can use them together. The Indeed connector is only in Claude for now, so this recipe is written for Claude. Lamr's own tools work the same way in ChatGPT.

Setting up takes a few minutes and you do it once. In Claude, open Settings, then Connectors. Add Indeed from the list and sign in. Add Lamr as a custom connector with the server address from the Lamr app's AI connections dialog, and approve the consent screen. From then on both are available in any new chat.

Step 1: search, and look before you save

Start with a search that is narrow enough to read. Ten results you actually look at beat a hundred you skim.

Search Indeed for [role] jobs in [location or remote], posted in the last week. Show me up to 10: title, company, location, pay if listed, and one line on what the role is. Don't save anything yet.

Then check what you already have, so you do not save the same role twice or reapply somewhere that turned you down last month:

Look at my applications in Lamr. Which of these companies have I already applied to or saved, and what happened?

Step 2: save the ones worth it, with the whole posting

This is the step that matters. Lamr treats the job description as a source record: fit scoring and tailoring read it line by line. A two-sentence summary gives you a vague score and a generic resume.

Lamr can import a description from a link on many hiring sites, but Indeed's pages block that kind of request. So ask Claude to pass along the full text it already got from Indeed:

Save numbers [2, 5 and 7] to Lamr as new applications. For each one, get the complete job description from Indeed and send all of it, with its headings and lists, not a summary. Include the posting link.

Saving is free. If a description comes through too short, Lamr says so instead of quietly accepting it, and you can paste the full text yourself.

Step 3: score your fit, then decide

Now screen what you saved. Each check costs 1 credit and compares the posting's requirements with your saved resume and profile, one requirement at a time.

Check my fit for each application I just saved. Give me the score, the requirements I meet, and the ones I'm missing. Tell me if any has a hard blocker like work authorization or location.

The scores are on one scale, so you can compare them. A common mistake is to tailor documents for everything you saved. Screen first, then spend the 5 credits on the two or three roles where your experience really matches.

Draft a resume and cover letter for [company], then compile them. Tell me what you changed and anything you couldn't support from my resume.

If your shortlist is long, ask for the weakest matches to be left as saved drafts. They cost nothing to keep, and a role that is a poor fit today can be worth a second look after you add a project or a skill to your master resume.

What this does not do

It does not apply for you. Claude gives you the posting link, and you open it and submit. When you have, say so, and the record stays right:

I applied to [company] today. Mark it applied and set a follow-up for a week from now.

It also does not replace reading the posting yourself. A fit score tells you whether your experience matches the requirements. It cannot tell you whether you want the job.