← All posts
Competition runEnveda CASMI 2026 · Kaggle · $50,000 · closes 14 Dec 2026

Enveda CASMI 2026 on Kaggle: From Rank 3,856 to 1,421, 0.008 From Bronze

Our Enveda CASMI 2026 run: 52 rounds, public score 0.136 to 0.479, rank 1,421 of 5,305 and 0.008 below the bronze-medal line. What worked, what failed.

1,421 / 5,305
Public rank (15 Nov)
0.008 from bronze
Medal standing
0.479
Best public score
52 · 10 rejected
Rounds

Manas Joshi·November 14, 2026·Updated November 15, 2026·8 min read

Enveda CASMI 2026: competition facts
  • Where: Kaggle (Featured), $50,000 prize pool, medals awarded
  • Task: Identify the 2D structure of unknown molecules from LC-MS/MS mass spectra
  • Score: MRR@25 over up to 25 ranked candidates per molecule
  • Dates: team merger 7 Dec 2026 · final deadline 14 Dec 2026
  • Field: 5,305 teams on the public leaderboard (15 Nov); leader 0.549
  • More: all Enveda CASMI 2026 facts, dates and FAQ
🗓️
Standing on 15 Nov 2026: rank 1,421 of 5,305 (top 26.8%) with a best public score of 0.479. The bronze-medal line is 0.487 (rank 530), so we are 0.008 below it with a month to go. This page is updated as the run moves.

The short version

ResearchForge has recorded 52 rounds on Enveda CASMI 2026: 18 in October from the research loop and 34 since 7 November from Kaggle Overnight, all 34 launched unattended by auto-run with Claude Code as the driver. Our own ideas took the public score from 0.136 to 0.213. The jump to 0.479 came from adopting a public pipeline whose author reported 0.500. We say that plainly: the record exists to show what actually moved the score.

Score timeline (public leaderboard)

DateLocal MRR@25PublicWhat changed
24 Oct0.89940.136Spectral library search only
25 Oct0.89940.156A fallback for molecules the library does not contain
29 Oct0.90320.158A second ranking signal, from top public notebooks
31 Oct0.89150.166Trusting the fallback more — local fell, public rose
1 Nov0.86490.164Fallback only for ranks 2–25 — pushed too far
11 Nov0.91150.181A different library similarity; both rose together
13 Nov0.91410.191Spectrum clean-up before matching
13 Nov0.91490.213A fragment-based score in the ranking — our own best
13 Nov—0.479A public pipeline adopted whole (reported 0.500)

The October and November local numbers come from two different benchmarks, so compare them within each month, not across.

localhost:7842
One experiment tree with October research-loop experiments dotted and November rounds solid
One tree: October’s research-loop experiments (dotted) and November’s overnight rounds

What worked

  • ▸A better library similarity. The single biggest step of our own: local 0.895 → 0.912 and public 0.166 → 0.181.
  • ▸Cleaning the spectrum before matching, then a fragment-based score on top: public 0.180 → 0.191 → 0.213.
  • ▸Building on strong public work. The largest gain by far, and the reason we are near the bronze line.
  • ▸Running unattended. Auto-run launched every November round on its own, so the loop kept going overnight and through the day.

What failed or was rejected, and why

RoundWhy it did not count
Fallback-only ranking (1 Nov)Local 0.865 and public 0.164 — both down; rejected
Top-2 spectrum aggregationLocal 0.894 → 0.873; not kept
Two gated variantsLocal 0.913 → 0.909; not kept
Two long roundsStill running on Kaggle when the 30-minute round limit hit; collected when they finish
Two rounds with no metric, one timeoutFinished without a score, or timed out — not counted as results
10 adopted public pipelinesRan to the end with no local score; marked unscored, only the leaderboard can judge them
localhost:7842
Enveda Runs page with adopted public pipelines marked unscored in purple
The dashboard’s Runs page: adopted public pipelines in purple, unscored and offered for submission (titles blurred while the competition is live)

The lessons

A local benchmark is only as good as what it holds out. In October ours held out spectra of compounds already in the library, so it could not see the fallback for unknown molecules, and the public score rose while local MRR@25 fell four times in a row. In November local and public moved together in five of seven steps. The full analysis is in Kaggle CV vs leaderboard.

Read the direction, not the size. A 0.0008 local gain came with +0.022 public, while two other sub-0.001 steps were ties. And a reported score is one draw: the pipeline reported at 0.500 scored 0.479 for us.

What’s next

Closing the 0.008 to the bronze line before 14 December, by combining the public pipeline with the ideas that held up on our own. Every round, kept or rejected, will be in the same record. See all four runs in our November standings, or run the loop on your own competition with Kaggle Overnight.

Try it on your next run

Every number in this post came from the ResearchForge record. Yours can too.