DoLoup writes the SAS program, runs it, reads the log, fixes the ERRORs — and loops. No babysitting, no manual log triage. When the loop ends, your dataset is built, QC'd against CDISC checklists, and ready to submit.
Speaks your standards
Every SAS programmer knows the loop: submit, open the log, hunt the ERROR, patch, resubmit. DoLoup turns that grind into an autonomous DO … UNTIL — an agent that iterates on your program the way you would, only without the coffee breaks.
Three steps between a specification document and a clean, submission-ready deliverable.
Describe the analysis in plain English, or attach DOCX and Excel specs. AI Scan extracts datasets, variables, and derivations automatically.
DoLoup drafts the full program, then enters its agentic loop: run in SAS, parse the real log, submit a targeted fix, rerun — hands-free until the log is clean.
One click on any dataset card generates an independent v_<dataset>.sas that checks every rule in your CDISC checklist and writes a findings CSV.
From first draft to double-programmed validation — without leaving the browser.
The signature loop: parse the real SAS log, detect ERRORs, WARNINGs, and notable NOTEs, submit an AI fix, rerun — autonomously, until clean.
Plain-English prompts become complete SAS 9.4 or R programs with your macro variables, LIBNAMEs, and house style built in.
Checklist-driven v_<dataset>.sas programs validate SDTM and ADaM datasets against your selected IG version, with Pass/Fail CSV reports.
A live diagram of programs, datasets, and outputs. Run programs, view data, or launch QC straight from any node.
Attach DOCX/XLSX specs to programs. AI Scan extracts metadata; Reconcile compares spec vs. produced datasets.
Program status, correction cycles, run history, and recent errors across every deliverable — in one view.
Git-backed history for every program with side-by-side diffs and one-click restore of any prior version.
Queue whole sections — SDTM, ADaM, TLF — with dependency ordering, or schedule overnight batch runs.
Role-based access, JWT sessions, per-user workspaces, and a full audit trail designed for regulated environments.
Right-click any SDTM or ADaM dataset in the Data Flow and choose Perform QC. DoLoup reads your CDISC checklist workbook, applies the IG version from your settings, and generates a fully independent validation program — written by a different AI model than the one that built production.
Real double programming means the second programmer isn't the first programmer wearing a different hat. DoLoup lets you assign separate AI models to Production and Validation — so the model that QCs your dataset never grades its own homework.
One workspace from raw data to tables, listings, and figures.
DoLoup runs on your infrastructure, next to your SAS installation. Set the agent looping — it will let you know when the log is clean.
Request a demo Watch the loop again