Monte Lattice

Job Engine

An AI-screened job-search pipeline with a live CRM, résumé compiler, and outcome tracking. Run daily. On my résumé this is the Document Automation Pipeline & Custom CRM.

Live from the Engine

Read at page load from the engine's public feed, as of Oct 18, 2026. Raw data →

These update themselves: the engine recomputes them every 15 minutes, and this page re-reads them at most every 5 minutes. A figure can be up to 20 minutes old.

4,595 Roles screened AI-evaluated, from 15,198 listings discovered
188 Applications sent logged in the CRM over 79 days
212 Live conversations open recruiter threads
22 Dead links retired of 41 detected · 13 this week

Pipeline volume by week

Where the pipeline stands

Listings discovered 15,198 Roles screened 4,595 Documents compiled since Sep 27, 2026 142 Applications sent 188

Each band is that stage's own running total, not a conversion from the one above: the counters cover different periods.

76 Median fit score out of 100, across 1,527 roles that cleared the screen
8 of 8 Résumé tracks tracks routed to so far
142 Documents compiled résumés and cover letters since Sep 27, 2026

Match to sendable draft

143.3 ms

resume PDF + cover letter PDF + email body, compiled for a real cached role. Median of 3 warm runs on the engine's own server, measured 2026-10-18 23:00:24 UTC.

The Gmail API round-trip that files the draft is network time and is not included.

What It Looks Like

A Telegram message showing a screened job card: role, company, fit score, contact and command buttons
A new listing arrives in Telegram as a card: the role, how well it fits, who to contact, and one-tap commands.
A Telegram message with a drafted outreach email and a tailored résumé PDF attached
One command later: the email is drafted in Gmail and a résumé tailored to the role is attached.
The Google Sheets tracker listing companies, roles, fit scores, status and follow-up dates
Every role lands in the tracking sheet with its status and next follow-up date.

How It Works

Source

Pull & screen

Listings are pulled from multiple boards and scored against a strict evidence bank. No invented experience.

Tailor

Compile outreach

A résumé and message are compiled per role, then queued in a swipe-to-send Telegram review.

Track

Log to CRM

Every send, reply, and interview lands in the CRM. The figures above are read from it and from the engine's own log.

Follow-Up Planner

Plan your follow-ups

These are the gaps the engine itself waits. Pick the day you sent an application or a first email, and the dates below are when it would follow up and close the loop.

  1. An application: follow up 10 days after you send it, close it out at day 21.
  2. A cold email: one nudge at day 4, then triage at day 21.
  3. Once someone replies: check back 4 days after the reply.
View source on GitHub →