Shivani Gupta’s latest article offers a behind-the-scenes look at her PharmaSUG 2025 experience and the launch of Submify. In her article, she explores how this AI-powered platform integrates CDISC CORE and automation to streamline submission workflows and improve data quality.
This year, I had the absolute joy of attending my first ever PharmaSUG in San Diego—and to make it even more exciting, I presented at it too! Not to boast but the room was packed to the point where the stream Chairs had to close the doors. It was my first time presenting at PharmaSUG and I was very excited to see my peers turning out for this hot topic. Integration of CORE in our platform resonated with people and there were a lot of questions and follow-up on CORE during and after the presentation. I would like to think it was all ME... but I must give credit to CDISC CORE – a free and open software built for the community to test study data for conformance to CDISC standards as well as regulatory and sponsor-specific conformance rule sets.
It was such a full-circle moment to finally meet, in person, the brilliant minds I’ve been virtually collaborating with for years. And yes, there were nerves but mostly, there was a deep sense of gratitude for the chance to learn, share, and be part of a community that genuinely cares about making clinical data better.
And at the heart of my presentation? Submify – our new AI-powered, standards-driven, open-source-loving platform designed to make regulatory submissions a whole lot smarter (and a lot less painful).
The Not-So-Glamorous Reality of Regulatory Submissions
Let’s be honest – submissions are very stressful.
- Validation happens too late.
- Workflows are pieced together like a patchwork quilt.
- Too much still depends on manual checks and commercial tools.
And the result? Frustration, delays, and that creeping fear of “what if we missed something?”
As someone who’s worked across different study teams, I’ve seen how painful these gaps can be. So, we asked ourselves: what if there was a way to bring it all together into one seamless, automated pipeline?
Meet Submify: A Smarter, Saner Way to Prep for Submission
Submify is our answer to submission chaos. It’s AI-enhanced, modular, and open-source ready—and it brings all the moving pieces of submission data prep into one beautifully orchestrated modular platform.
1. CORE Compliance (Without the Constant Headaches)
We’ve embedded the CDISC Open Rules Engine (CORE) right into the workflow so that:
- Conformance checks run early and often (not just before the finish line)
- Both standard and custom (study-specific/sponsor specific) rules can be created and applied
- AI flags repeat offenders and suggests resolutions
- Outputs flow directly into define.xml and reviewer guides
It's like having a hyper-organized friend who color-codes your compliance issues before they become problems – and the compliance checks are free when using CORE!
A standout resource in this area is the PharmaSUG 2025 paper by my new friend Lex Jenson,
titled "Compliance Checks: Running the CDISC Open Rules Engine (CORE) in BASE SAS©." His
work provides a practical and scalable approach to integrating CORE with BASE SAS.
We’ve referenced Lex’s approach in Submify for the CORE Compliance Checks.
2. Auto-Generated Metadata & Deliverables (Yes, Really)
Say goodbye to the hours spent checking the define.xml, making sure your aCRF annotations are done correctly or trying to create bookmarks for aCRFs.
Submify:
- Creates key artifacts: define.xml, cSDRG, ADRG, bookmarks to aCRF
- Checks the define.xml to make sure there are no broken hyperlinks and annotations are done correctly
- Uses AI to generate context-aware reviewer guides
- Ensures consistency across datasets, metadata, and validations
It’s the kind of automation that saves time—and a little sanity.
3. eCTD-Ready Without the Last-Minute Panic
In Submify, we have built Python scripts to validate:
- Folder structure
- File names and formats (.xpt, lowercase enforcement, etc.)
- Inclusion and correct placement of every required document
AI even catches missing files or odd mismatches based on what it's learned from past submissions. So, when it’s time to package up for eCTD, everything’s already in the right place.
4. Built for Open-Source Collaboration
Because good science is collaborative, Submify:
- Works with CDISC CORE, CDISC 360i, and COSA tools
- Offers plug-and-play modules for sponsor-specific workflows
- Keeps everything version-controlled and audit-ready
And best of all? No gatekeeping. Anyone on your team can contribute—without waiting for that one person with the commercial tool license.
Why It Matters: The Real Impact
We’re still in the early stages, but here’s what we’ve seen so far:
- Faster timelines – Up to 50% reduction in submission prep time
- Stronger data quality – Thanks to continuous validation
- Better collaboration – Open tools, not fragmented workflow
- Risk mitigation – No more waiting till the end to find issues
- Lower costs – Less rework, more efficient use of team time
What’s Next?
Submify is evolving fast—and the future looks bright. We’re continuing to explore ways to make the platform more intuitive, integrate even more AI logic, and build visual interfaces so non-programmers can be a part of the workflow too.
For me, this project is about more than just compliance. It’s about giving teams the tools they need to work smarter, stay aligned, and focus on what really matters—delivering high-quality, submission-ready data with confidence.
If you’re curious about Submify, CORE, or anything in the CDISC + AI + automation universe.
Let’s connect - sgupta@clymbclinical.com
Director of Statistical Programming
Latest Blogs
By Malan Bosman
April 30, 2025
TFL Designer + ARS = TFL Automation!
At PHUSE US Connect 2025, Malan Bosman shared how Clymb Clinical is automating the way TFLs (Tables, Figures, and Listings) are created. In this article, Malan walks through our step-by-step workflow using TFL Designer to build shells and capture metadata, CDISC’s ARS standard to structure results, and the siera R package to auto-generate R scripts. Check out how Clymb Clinical turn it all into polished reports (RTF/PDF) with just a few clicks. If you're in stats, programming, or clinical data, this is a must-read.
By Clymb Clinical
February 25, 2025
Clymb Clinical & CDISC: Transforming Clinical Data Standards with the eTFL Portal
At Clymb Clinical, we’re passionate about making clinical research more efficient, automated, and standardized. That’s why we’ve collaborated with CDISC to launch the first ARS-compliant packages in the eTFL Portal—a game-changing step in modernizing how analysis results are managed and shared.
By Malan Bosman
November 18, 2024
From Shells to TFLs: Leveraging Industrial Engineering Principles for Automated and Efficient Clinical Data Outputs
As an Industrial Engineer working as Statistical Programmer in the CRO industry, I’ve often been asked the question (by others as well as myself), “How did you end up here?” The role of Statistical Programmer has the reputation of being defined by rigid rules and regulations, following strict industry guidelines to produce pre-defined dataset structures and Tables, Figures, and Listings (TFLs).
By Navin Dedhia
September 25, 2024
Startup Innovation and Growth Through Microsoft Founders Hub
At Clymb Clinical, we’re committed to innovating and improving clinical data analysis and reporting by building our own software solutions and integrating existing technology stacks, including available open-source resources. As a startup, scaling efficiently is key, but as you scale you also have to carefully manage costs to sustain growth.
By Anna Yaggi, Clymb Clinical
September 19, 2024
Expect the Unexpected: Working at a Data Services Start-Up
During my undergrad, all I knew was that when I joined the workforce, I wanted to use my degree for good and work somewhere where I felt like I was making a difference. I didn’t want to get lost in a large corporate environment, but I didn’t necessarily think of myself as someone who would work at a start-up.
