Recent Identity.md ($IMD) posts on X
Researched 7 November 2026, through 17:44 UTC.
These are the ten newest qualifying posts I could substantiate in the public-source sample below. They are not a certified platform-wide latest ten. Every listed author had X's is_blue_verified: true and at least 10,000 followers when checked. Eight entries are replies, one is an original post, and one is a quote post. Brief reactions are included because the request asks for recent posts, rather than only substantive analysis.
“Top” is interpreted here as 10,000+ followers, an analyst-selected audience threshold, not an X ranking or a measure of expertise. “Verified” means the observed blue-check metadata; it does not establish the author's identity, independence, credibility, or agreement with IMD's claims. Counts are current observations, not follower counts at publication.
Ten newest observed posts, newest first
All times are UTC. Post links support the attribution and timestamp; linked follower counts lead to the corresponding X profiles. Summaries describe what the author posted, not independently established project results.
| # | Published (UTC) | Author · observed followers | Type | What was posted and why it concerns IMD |
|---|
| 1 | Nov 7, 16:32:31 | @Rumii_404 · 87,200 | Reply | “first place has a nice ring” — a brief reaction in Taylor's IMD trading-bot challenge thread. Post |
| 2 | Nov 7, 16:10:18 | @AuraBytes0es · 11,635 | Reply | Asks, “What made their strategy different?” in that same IMD challenge thread. This is a question, not evidence of a particular strategy. Post |
| 3 | Nov 7, 15:54:29 | @taylor_ · 18,453 | Original | Introduces an explanation of how imd.fun earned first place in Paradigm's trading-bot challenge. The expanded URL explicitly identifies the project. Post |
| 4 | Nov 7, 02:07:03 | @tangsanzangbtc · 13,009 | Reply | “ai agents figuring out trades without us is wild” — reacting to Taylor's IMD first-place announcement. This is commentary on the announcement, not an independent test. Post |
| 5 | Nov 7, 01:45:36 | @taylor_ · 18,453 | Reply | “the swarm is warming” — a short follow-up within the IMD first-place thread. It adds no measurable performance evidence. Post |
| 6 | Nov 7, 00:59:25 | @surfcoderepeat · 24,679 | Reply | Says he is testing and improving the goal feature, linking directly to imd.fun. This reports work in progress. Post |
| 7 | Nov 6, 21:02:27 | @taylor_ · 18,453 | Reply | Proposes putting @mpp in front of the swarm so other agents can hire it. This is a suggestion in the IMD thread, not a shipped-integration announcement. Post |
| 8 | Nov 6, 20:25:54 | @cheuk_baby · 15,885 | Reply | “看起来竞争挺胶着的啊” — my translation: “The competition looks pretty close.” Replies to Adam's IMD second-place update. Post |
| 9 | Nov 6, 20:24:29 | @PantherQB19 · 41,824 | Reply | Asks whether their agents can help, replying to Adam's IMD competition update. This expresses interest in participating; it does not demonstrate participation. Post |
| 10 | Nov 6, 19:28:02 | @0xminion · 28,702 | Quote | “OoooO? #1 now” — quotes their earlier post explicitly discussing the IMD swarm's /goal work on Paradigm's prediction-market challenge. Post, quoted post. |
Context behind the reactions
Taylor's first-place announcement claims that 202 IMD agents reached first of 188 players in 21 attempts, at a total cost of $14.52, without a human writing the strategy. That is Taylor's claim, not a benchmark reproduced in this research. The initial post was published Nov 6 at 19:26:02 UTC; its edited version is timestamped 19:28:58 UTC. These are versions of one announcement, not two separate posts. Initial version, edited version.
Adam's preceding update says the IMD swarm was second and that they were improving the goal feature. He speculates that a competitor might benefit from their publicly visible code; he explicitly expresses uncertainty. This report does not treat that speculation as an observed event. Adam's update, Nov 6, 18:57:17 UTC.
The clustering around these threads suggests the visible discussion was concentrated on the goal feature and the competition result. That is an inference about this sample, not a measure of overall X sentiment. Several replies are too brief to support conclusions about adoption, investment conviction, or technical understanding. The directly relevant original explanation is entry 3; many of the newer entries are reactions to it or to the preceding announcement.
How the evidence was checked
- Discovery used public web searches for
$IMD, Identity-md, Identity.md, imd.fun, and associated handles; indexed mirrors supplied additional post links and reply leads. Unrelated weather-department and other “IMD” results were excluded.
- Public X profiles were inspected for the initial candidates @surfcoderepeat, @taylor_, @blknoiz06, @0xminion, @yadfarms, @theneetguy, @Bankless, @Eli5defi, @AdamOnFinance, @wegocrypto88, @notwashed, and @nftimm. Reply discovery expanded the candidate set. This was a bounded discovery sample, not an exhaustive account census.
- For every table entry, the individual X post and its author's X profile were retrieved directly. The public server-rendered data supplied
Tweet.details.created_at_ms, User.verification.is_blue_verified, and User.relationship_counts.followers. Reply or quote context was checked to establish relevance even where the author's own words do not contain “IMD.” Follower counts in the table come from the author's own profile response.
- Entries are ordered by publication time, not likes, follower count, mirror-relative age, or repost time. Multiple posts by one author are allowed. Bare reposts, mention-only replies, and emoji-only replies were excluded. Edited versions count once.
- The compact extracted evidence, source URLs, field names, exact timestamps, short excerpts, and context post IDs are delivered in
sources/evidence.json. It is a research extraction, not a signed X archive. No dependency or network access is needed to read the deliverable.
Uncertainty and unanswered questions
The ten entries, timestamps, observed audience counts, and blue-check fields were substantiated from X's public responses. Completeness remains unverified: public search did not expose a usable live results feed, public profiles exposed only limited timeline samples, and the full reply trees were not paginated. Newer qualifying posts may exist outside the retrieved sample. Mirror caches also differed, so their relative ages were not used for final ordering.
Whether these are the ten most recent across all qualifying X accounts remains unanswered. The request also does not define “top”; a threshold of 100,000 followers, an engagement ranking, or exclusion of replies would produce a different list. This report makes its 10,000-follower interpretation explicit rather than presenting it as the requester's choice.
The claimed leaderboard results, agent autonomy, spending, and project adoption were not independently audited. Account relationships, sponsorship, and coordination were not established. A blue check and a large audience do not resolve those questions. No claim is made that these are ten independent endorsements or ten different authors.