Arctic Shift is an 🏷️#open-source project by Arthur Heitmann that makes large-scale 📝Reddit data available to researchers, 📝Moderators, and developers through three surfaces: bulk data dumps, a public 📝API, and a web search interface.
What Does Arctic Shift Offer?
- Bulk data dumps — compressed archives of Reddit posts and comments, with helper scripts for efficient processing of multi-gigabyte files, plus a
/download-toolpage for pulling a single user or a smaller subreddit without touching the full dumps - A public API — programmatic access to historical posts, comments, and user activity without scraping
- A web search interface — a lightweight search layer over the same archive at
/search, for anyone who needs a specific post or comment rather than bulk access - A removal-request mechanism — any user can request removal of their own content from the archive
How Does the Arctic Shift API Work?
The API serves the same archive as the bulk dumps over plain unauthenticated HTTP — no key, no registration — which matters because Reddit's tightening API policies have made independent data collection harder for non-commercial users. The base URL is https://arctic-shift.photon-reddit.com, the search UI sits at /search, and Heitmann offers no uptime or performance guarantees.
/api/posts/searchand/api/comments/search— filter byauthor,subreddit,after,before, andsort, withlimitfrom 1-100 orautofor 100-1000 results. Posts addtitle,selftext,query, andurl; comments addbody,link_id, andparent_id./api/posts/ids,/api/comments/ids,/api/subreddits/ids,/api/users/ids— direct lookup by comma-separated IDs, capped at 500 per request./api/comments/tree— the full comment tree beneath a post vialink_id, up to 25,000 comments./api/posts/search/aggregateand/api/comments/search/aggregate— counts grouped bycreated_utc,author, orsubreddit./api/subreddits/search,/api/subreddits/rules, and/api/subreddits/wikis— subreddit metadata, rule sets, and wiki pages./api/users/searchand/api/users/interactions/users— user lookup and user-to-user interaction graphs.
Keyword parameters — title, selftext, body, and query — only work alongside an author or subreddit, and are unsupported for very active users or subreddits. Every endpoint accepts fields to trim the response payload; the search endpoints also take format as json, rss, or jsonfeed.
curl "https://arctic-shift.photon-reddit.com/api/posts/search?subreddit=worldnews&limit=2&fields=id,title"The bulk dumps expose the same archive for large-scale analysis, and the project's helper scripts support efficient handling of Reddit's compressed data formats.
Why Does the Removal-Request Mechanism Matter?
Anyone can request removal of their own content through a public request form linked from the project's repository, and can first confirm whether their data is in the archive by searching their username in the search UI. This governance layer addresses an often-underappreciated dimension of large-scale social-data hosting and distinguishes Arctic Shift from purely scraped datasets that bypass user consent entirely.
How Does Arctic Shift Compare to Other Reddit Research Tools?
For the broader cluster of research tools, see 📝Reddit Research Tools. For the full tool landscape, see 📝Reddit Tools. For Reddit's commercial data licensing path — the contrast to Arctic Shift's research-focused archive — see 📝Reddit's Data Licensing Deals.
Arctic Shift aligns with the kind of foundational tech we build at a venture studio — bulk dumps for deep analytics, API endpoints for rapid prototyping, and a deliberate governance posture on removal requests. A strong building block for anyone exploring social-data research, moderation tooling, or community-insight verticals.