exametrika
A test-theory package that grew into a graphical-model toolkit, now spending its releases paying down the API debt that growth created.
A side-by-side editorial comparison of inti and Statsig — release velocity, themes, recent moves, and the top alternatives to consider.
inti keeps compounding small statistics and publishing tools for plant-science labs.
inti bundles the tooling a plant-science lab uses end to end: the Yupana analysis app, the Tarpuy experiment planner, scihub and rticle document rendering, and helpers for ANOVA and heritability. The 0.7.x line has been dominated by one thread, a PCA sub-module in Yupana carrying variable contribution, dimension correlation, supplementary variables and per-dimension selection. Releases are cumulative: 0.7.0, 0.7.1 and 0.7.2 all restate the same PCA lines, with each tag adding a few items on top, and 0.7.0 itself is split across two tags published 35 seconds apart.
Statsig is packaging its own workflows as skills for someone else's agent to run.
Statsig's experimentation and feature-flag platform opened a public agent-skills repository holding reusable workflows — create dashboard, create cloud metric — written as instructions an AI agent can execute. Its MCP server extended to Segments and Layers, covering user targeting and experiment configuration, and Metrics Explorer gained the ability to abort long-running queries. The crawled feed captures marketing pages as entries, so several rows carry navigation boilerplate rather than release notes.
inti bundles the tooling a plant-science lab uses end to end: the Yupana analysis app, the Tarpuy experiment planner, scihub and rticle document rendering, and helpers for ANOVA and heritability. The 0.7.x line has been dominated by one thread, a PCA sub-module in Yupana carrying variable contribution, dimension correlation, supplementary variables and per-dimension selection. Releases are cumulative: 0.7.0, 0.7.1 and 0.7.2 all restate the same PCA lines, with each tag adding a few items on top, and 0.7.0 itself is split across two tags published 35 seconds apart.
Two axes are moving. Analysis is deepening inside Yupana, where PCA went from a single view to a sub-module with its own contribution and correlation outputs across three tags. Publishing is widening around rticle() and scihub(), which now handle Google Docs markdown, crossrefs and page numbers, continuing the gdocs2qmd work from the 0.6 line. Neither is a change of direction; the package accretes features where the maintainer's own research workflow needs them.
Expect the next tag to extend the PCA sub-module again and add another rticle() or scihub() rendering detail, on the two-to-six-week cadence the 0.7 line has held.
Statsig's experimentation and feature-flag platform opened a public agent-skills repository holding reusable workflows — create dashboard, create cloud metric — written as instructions an AI agent can execute. Its MCP server extended to Segments and Layers, covering user targeting and experiment configuration, and Metrics Explorer gained the ability to abort long-running queries. The crawled feed captures marketing pages as entries, so several rows carry navigation boilerplate rather than release notes.
The direction is agent-operated experimentation. MCP covering segments and layers means an agent can configure targeting, and the skills repository means the multi-step workflows around that configuration are distributable artifacts rather than documentation. Statsig is betting the console is not where experiments get set up much longer.
Expect the skills catalog to grow toward the analysis side — reading experiment results and proposing rollout decisions, not just creating objects. The query-abort work suggests warehouse cost control is a live concern as agent-driven usage increases.
Other Infra & APIs products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either inti or Statsig.
A test-theory package that grew into a graphical-model toolkit, now spending its releases paying down the API debt that growth created.
nuggets keeps compounding on the 2.0 rewrite — more pattern families, lighter install.
projoint spent a year on CRAN paperwork, then shipped a correctness fix it flagged itself.
eratosthenes spends 0.1.0 hardening inputs rather than adding chronology methods.
dqcheckr adds drift analysis, then removes the YAML a user had to hand-write.
An actuarial mainstay spends its releases on CI plumbing, not on new mathematics.
See all inti alternatives → · See all Statsig alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. inti is currently shipping more aggressively (velocity 2.5 vs 0.0), with 0 editorial sparks in the last 30 days against 0. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. inti is currently shipping more aggressively (velocity 2.5 vs 0.0), with 0 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Infra & APIs products to evaluate alongside.
Top inti alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "inti alternatives" section above for the current picks, or visit /alternatives/inti for the full list with editorial commentary on each.
Top Statsig alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "Statsig alternatives" section above for the current picks, or visit /alternatives/statsig for the full list with editorial commentary on each.