RCode.ai
The AI copilot that actually writes R — for analysts everywhere.
tidyverse, ggplot2, Shiny, and statistics. Paste your code, ask anything, get runnable R back — not a Python answer with an "in R it's similar" footnote.
Free forever · 30 messages/day · No credit card required
Features
Built for how R is really used
One assistant covering the whole R workflow: wrangling, plotting, modeling, and apps.
Fluent in tidyverse
dplyr pipelines, tidyr reshaping, purrr functionals — answers written the way R is written today, with data.table or base R when that's what your code uses.
ggplot2 that runs
Complete, minimal examples with aes, geoms, scales, and themes. Copy, paste, and see the plot — no pseudo-code fragments.
Shiny debug mode
Reactive graph confusion, invalidation loops, observers that never fire — a mode built for finding out why your app does that.
Statistics, done properly
lm, glm, mixed models, survival analysis, and pharma reporting conventions. Explanations a reviewer would accept.
Context from your code
Paste your script next to the chat. Every answer is grounded in the code you're actually working on, not a generic example.
Your account, your limits
Simple accounts with transparent daily usage. No surprise bills, no training on your code.
Pricing
Start free, upgrade when it earns it
The free tier is a real tier, not a demo — for individuals and teams in any timezone.
Pro
$12 /month
- ✓ Unlimited messages
- ✓ Shiny debug mode
- ✓ Saved session history
- ✓ Early access to the IDE extension
Team
$29 /user/month
- ✓ Everything in Pro
- ✓ Per-seat Stripe billing
- ✓ Invoice-ready billing via Stripe
- ✓ Contact us for shared workspace (roadmap)
FAQ
Questions R users ask
General assistants default to Python and produce R code that looks plausible but mixes idioms from three ecosystems. RCode is prompted, tuned, and tested for R: it follows your style (tidyverse, data.table, or base), and it never answers an R question with 'in Python you would…'.
No. RCode sees only the code and messages you paste into the chat. Nothing is used for model training.
The tidyverse, ggplot2, Shiny, statistical modeling (lm/glm, mixed models, survival), renv and package management, and biostatistics reporting conventions.
A VS Code / Positron extension is on the roadmap. Pro subscribers get early access.
Stop translating from Python
Ask your next R question somewhere that speaks R — from any lab, any country.
Open the playground