SimplifyJobs/New-Grad-Positions: A Community Job Board for 2026 Entry-Level SWE, Quant and PM Roles

What New-Grad-Positions actually is, and who it is for

This repository is not a program you install. It is a README file that holds a set of HTML tables, one per job category, plus a legend and a contribution guide. The README describes it as a place to "share and keep track of entry-level software, tech, CS, PM, quant jobs for 2025 & 2026 new graduates," maintained by Coder Quad and Simplify. The audience is narrow and clearly stated: people finishing a degree in 2025 or 2026 and applying to their first full-time role.

The README groups openings into six categories with counts in the headings: Software Engineering (233), Product Management (6), Data Science, AI & Machine Learning (85), Quantitative Finance (18), Hardware Engineering (70), and Other (0). The introduction says the list holds 412 roles in total. Each row carries a company, a role title, one or more locations, application links, and an age in days. A legend defines five markers that appear next to entries: no sponsorship offered, US citizenship required, application closed, FAANG+ company, and advanced degree required.

That is the whole product. There is no search, no filter, no notification system inside the repository. If you want alerts, the README points to swelist.com, which it says a community member built to email new roles from this repo. The repository's own homepage field also points at swelist.com, so the project's public face is a third-party site rather than anything shipped from this codebase.

How the table is built and what the Age column really means

The mechanism is a README that a bot rewrites. A comment in the file reads "Please leave a one line gap between this and the table TABLE_START (DO NOT CHANGE THIS LINE)", which tells you an automated process replaces everything below that marker with freshly generated tables. The contribution path runs the other way: the README asks readers to submit an issue using the repository's issue template and points to CONTRIBUTING.md for the rules.

So the data flow is: an employer posts a job, a scraper or a human notices it, the table is regenerated, and the row appears with an age of 0d. The Age column counts days since the row entered the list, not days since the employer opened the posting. Those are different numbers, and the README does not claim otherwise.

Every application link is wrapped in tracking parameters. Company links carry utm_source=GHList&utm_medium=company, and direct application links carry utm_source=Simplify&ref=Simplify. Each row typically offers two links: one to the employer's own applicant tracking system (Workday, iCIMS, Ashby, Greenhouse-style hosts appear in the sample rows) and one to a simplify.jobs/p/ URL. If you care about not being tracked, use the employer link, and expect to strip the query string by hand.

The repository's top level contains only .github/, CONTRIBUTING.md, README.md and archived/. That archived/ directory is the interesting one: it implies rows are retired rather than deleted, so the current README is a working set and the archive is the history. The README does not explain what triggers a move to archived/ or how to read that directory.

Reading the README and getting to a first usable shortlist

There is nothing to install. The project is consumed by cloning it or reading it on GitHub, and the README gives no installation steps because none exist. The README's contribution section is the only workflow it documents, and it points readers at the issue template rather than at any command:

https://github.com/SimplifyJobs/New-Grad-Positions/issues/new/choose

That URL is the repository's own contribution link, reproduced from the README. Opening it in a browser is how the project expects new rows and corrections to arrive; there is no CLI, no package and no build step to run first. The default branch is dev, so if you do clone the repository, that is the branch the maintainers write to, and the entire dataset is the single README.md at its root.

Once you have the file, open it in an editor with a markdown preview so the HTML tables render, and work down the Application column. The tables are plain HTML inside markdown, so grepping for rows is the only local processing the format supports, and the README does not document any parser or script for it. When a row looks interesting, click through to the employer's own posting and confirm it is still open before you spend time on an application.

If you want alerts instead of manual checking, the README directs you to swelist.com, described there as sending email alerts for new roles in this repo. That is a separate service, not part of the repository, and the README does not document how it authenticates or what it stores.

Where this list breaks down

The first limitation is that a README is not a database. There is no schema, no API, no JSON export, and no versioning of individual rows. If you want to build a dashboard, a tracker, or an alerting pipeline of your own, you are parsing HTML tables out of markdown, and the TABLE_START marker warns you that the format is machine-generated and not a stable contract.

Second, the categories are fixed and the list is US-centric. Every location in the visible rows is a US city or "Remote in USA." The legend's sponsorship and citizenship markers (no sponsorship, US citizenship required) only make sense against US work authorization rules. A graduate applying in Canada, the UK or India gets a table that mostly does not apply, which is worth knowing before you invest an evening in it.

Third, the Age column is a weak freshness signal. A row at 0d means the row is new to the list, not that the employer is still accepting applications. The legend has a marker for closed applications, which implies closures are detected and flagged, but the README does not say how often that check runs or how stale a row can get before it is corrected. Dead links are a real possibility, and the only defense is the employer's own page.

Fourth, the numbers in the headings are a snapshot written into the file. The introduction says 412 roles, the Software Engineering heading says 233, and Product Management says 6. If the bot regenerates the tables but not the prose, those two numbers can drift apart, and there is no changelog to tell you which is current.

How it compares with a general job board

The obvious alternative is a general board like LinkedIn, Indeed or a university career portal. The difference is not coverage, it is the shape of the filter. A general board starts with everything and makes you narrow down; this repository starts with a hand-checked category list and a set of markers that a general board does not expose as first-class fields. The no-sponsorship and citizenship-required flags are the clearest example: on a general board you find that information by reading the posting, and on this list it is a symbol in the row.

A second alternative is a structured aggregator with an API, where you can query by date, location or keyword. This repository cannot do that. It trades query power for editorial judgment about what counts as an entry-level role in a defined set of fields, and for a list small enough to read in one sitting. That trade is reasonable if you are a new grad in the US applying broadly and want to see the whole board at once. It is a bad trade if you have a narrow search and want to be notified the moment a specific company posts.

The repository also sits inside an ecosystem it does not control. The maintainers are Coder Quad and Simplify, a company that sells an application-autofill extension, and the README promotes that extension in a banner. The list is free to read, but the surrounding calls to action are commercial. That does not make the rows wrong; it does mean the repository is a marketing surface as well as a dataset, and you should read the company links with that in mind.

Maintenance, licensing and what to verify before you rely on it

The README states the list is "updated and maintained daily by Coder Quad and Simplify," and the presence of 0d and 1d ages in the sample rows is consistent with frequent regeneration. The repository is not archived. Beyond that, no last push date, release history or changelog is given, so there is no way to confirm the daily claim independently from the README's own sentence. Treat the update cadence as a claim to check, not a fact.

The licence is not stated anywhere in the repository files. There is no LICENSE file at the top level, and the README does not name a licence. That matters if you want to reuse the data: without a licence, the default position under most copyright regimes is that no reuse rights are granted, and republishing the tables elsewhere is legally unclear. This is not legal advice, and if you plan to build on the data you should ask the maintainers directly rather than assume.

The upgrade cost is close to zero, which is the honest upside of a README-based project. There is no dependency to update, no breaking change to track, no migration. The cost you pay instead is manual: every time you want current data you re-read the file, and every row you act on needs a click through to the employer to confirm it is live. There is no maintenance burden on you, and also no guarantee on the other side.

Editorial conclusion

Use it if you are a 2025 or 2026 graduate hunting entry-level SWE, quant, PM, data or hardware roles and you want one page with company, title, location and an application link. Skip it if you need structured data, an API, or roles outside the categories the README carries, and skip it entirely if you are not in the United States, since the sponsorship and citizenship markers assume a US hiring context. Before relying on it, open the archived/ directory to see what has already been retired, check the Age column on the rows you care about, and confirm the posting is still live at the employer's own URL rather than the Simplify redirect. The repository is a README with a table in it; treat every row as a lead to verify, not as a listing you can trust.

Frequently asked questions

Official sources

1. Official documentation
2. Official README
3. Project repository