In this repo, I share my scripts for Extraordinary Ability Green Card (EB-1A/EB-1B) applications.
One important step of the EB-1A and EB-1B application process is obtaining recommendation letters from independent recommenders—people you don't know personally and have never coauthored papers with. While some people attend conferences to connect with potential recommenders, I chose to reach out directly to researchers who have cited my papers by sending cold emails requesting recommendation letters. To get started, you need to compile a list of potential recommenders and gather content for your cold email outreach. Here, I would like to share some scripts I used for this process. I would like to share my cold call statistics: I sent 94 cold call emails for my EB-1A application. Eight researchers rejected my requests, nine researchers accepted them, and the remaining 77 researchers did not reply. I lost the detailed cold call statistics for my EB-1B application since I was using my work email.
The next step involves preparing a detailed petition letter that showcases your research accomplishments.
I modified Graham Neubig's research career tools repo to generate a CSV file for conducting further data manipulation.
The CSV file contains the following columns:
- Rank
- Citing Author
- Author ID
- Total Citations
- Author Paper Count
- Cited Papers and Citing Papers
Based on this information, I can identify recommenders who best fit my EB-1B and EB-1A applications, particularly those with high paper counts who have either cited my papers multiple times or cited multiple papers of mine.
Using the CSV file, I can generate cold call emails with detailed information about each potential recommender.
Additional information such as citer names, emails, and locations can be added by modifying the script to call other APIs. Location data is particularly important for EB1A applications, which require demonstrating global recognition through citations from different continents.
Use AI to analyze citation highlights and determine whether they are positive or negative. Based on this enriched information, filter the list of citers.
While the CSV file can be used to create an AI agent for automatically generating and sending cold call emails to potential recommenders, I took a different approach for my applications. I chose to review each recommender's profile individually and add personal touches to each email.
However, if you're short on time and want to reach out to as many recommenders as possible, you can leverage the CSV file with an AI agent or workflow to automate email generation.