RESiQ turns the evaluations you already collect into one clear record per resident.
RESiQ reads resident evaluations in any format and gives reviewers a structured summary of each resident before the meeting. If your program uses a system like MedHub or New Innovations, RESiQ works alongside it.
From a stack of forms to one record per resident.
RESiQ works with the evaluations you already have, whatever format they are in. It does the compiling, so reviewers can spend their time on the residents.
-
Ingest
Reads PDFs (native and scanned), images, Excel, HTML and Word exports. Handwritten forms are read with a vision model.
-
Classify
Recognizes the type of each form, so form type is not a limiting factor.
-
Extract
Pulls scores and comments into structured fields, with tiered retries for difficult forms.
-
Trace
Logs every stage. Each value carries a reference code back to its source form, so reviewers can check it.
-
Deliver
Produces one structured Excel workbook per resident, ready before the review meeting.
See RESiQ run.
A short walkthrough: evaluations in mixed formats go in, and one workbook per resident comes out.
Demo video coming soon
Sample output
The sample workbook uses fictional residents, faculty and programs.
Early results from an earlier version
Every score and comment links back to its source form, so reviewers can verify what was extracted.
Based on early use of an earlier version of the tool in one program at one organization. The current version of RESiQ is being built to work across organizations.
What RESiQ is built to support.
Review prep is where we started. The same structured data is built to support more of what programs are asked to do.
Listed from what works today to what we are still testing.
-
Faster review prep
Compile and summarize a resident's evaluations before the semi-annual Clinical Competency Committee review.
Used in an earlier version -
Standardization across programs
One consistent structure and centralized reporting across programs that use different forms and systems.
Validating with programs -
Accreditation readiness
Consistent, traceable evidence of how residents are evaluated across the ACGME core competencies.
Exploring -
Compliance visibility
Spot gaps in evaluation completeness and consistency before they become reporting problems.
Exploring -
Retention and development
A longitudinal record that shows what shapes high performers and who may be at risk of leaving.
Early hypothesis
Where we are, and where we are going.
A record that follows every physician from their first evaluation to their last.
Over time that record can help programs and institutions see what shapes strong physicians, and build environments that keep them.
Today
We are hereWhat exists now
- Excel summary for each resident
- Ingestion of any format, including handwritten scans
- An earlier version used in one Internal Medicine program
Next
Planned next
- A live dashboard that fills as evaluations arrive
- HIPAA-compliant infrastructure
- Direct integrations with MedHub and New Innovations
Later
Where it leads
- Longitudinal physician record
- Retention risk and recruitment insight
- Cross-program benchmarking
- Predictive recommendations
This reflects our current plans. It will change as programs tell us what matters most.
How RESiQ handles your data.
How the tool works today, what we recommend for pilots, and what is not in place yet.
Where data goes today
- RESiQ runs as an application on a computer your program controls. Evaluation files and the workbooks RESiQ produces are stored there, not on Sabrati servers.
- To read and extract the forms, RESiQ sends their content to Anthropic's Claude API. That is the only external service the tool uses, and the only point where data leaves the computer.
- Under Anthropic's commercial API terms, data sent through the API is not used to train its models.
- Sabrati does not receive a program's evaluations or workbooks unless the program chooses to share them with us, for example to troubleshoot a problem.
- Because files stay on your computer, your program controls how long they are kept and when they are deleted. Retention of data sent to the API is governed by Anthropic's terms.
What we recommend for pilots
- Remove resident names and other identifiers from evaluations before processing. RESiQ reads forms as they are and does not remove identifiers for you.
- Check comments for patient details and remove them too. Narrative comments can sometimes contain them.
- Talk with your institution's privacy and compliance offices before processing evaluations. Resident evaluations are education records, and your institution decides what may be shared with a service provider.
What is not in place yet
- HIPAA-compliant infrastructure is on our roadmap and is not built yet.
- We are not able to sign a business associate agreement today. That is why we recommend that pilot data have identifiers removed.
- We are open to a data use agreement and to your institution's security review before a pilot starts.
- If we add hosted features, we will update this page before they are offered to a pilot program.
Frequently asked questions.
Answers to the questions programs are most likely to have.
How do reviewers know the data is right?
- Every comment is captured word for word, one entry per respondent. Nothing is paraphrased or shortened.
- Each comment is checked against its source document. If it cannot be matched, it is flagged to check against the original, never dropped.
- A second pass looks for questions that were missed. Anything still missing is flagged for human review in a plain-language report.
- A question that is on a form but has no answer shows as "NO DATA," not a blank cell.
- Every step is logged, so any value can be traced back to the form it came from.
- When comment summaries are turned on, they never soften criticism. Critical and recurring issues are flagged, and each summary cites the comments behind it.
- RESiQ prepares the record. Reviewers still read it and make the decisions.
Has RESiQ been used in a live program?
- An earlier version of the tool was used in one Internal Medicine program at one organization. That is where the early results on this page come from.
- The current version is being built to work across organizations, and that work is still in progress. Pilots of the current version have not started.
What does a pilot involve?
- Scope. A single program.
- Timing. One milestone evaluation cycle for each resident in that program. Milestone reviews happen twice a year, so a pilot covers one of those review periods.
- Start date. There is no timeline yet for when pilots can start. For now we are only collecting letters of support while we continue developing the systems.
- Letter of support. When a program signs up for a pilot, we collect a letter of support. We use it to help secure funding that lets us build out the technical stack and establish the requirements for organizational use in an early pilot.
Does a pilot cost anything?
- No. Pilots are free. Programs can use RESiQ for at least one review cycle at no cost.
- We may extend a pilot to collect more feedback.
- Early adopters will have the option to sign on to the product at early supporter rates.
Which formats work?
- RESiQ reads evaluations in the formats programs already have: digital and scanned PDFs, handwritten forms, Excel, HTML, Word documents and images.
- Form type is not a limiting factor. RESiQ reads the evaluation forms your program already uses.
- How data is handled, including privacy, is described in How RESiQ handles your data.
Our focus right now: discovery and new pilot programs.
We are early, and we want to build this with the programs that use it.
Discovery conversations
We are learning how evaluation and review work in your program today, and where the process is hardest.
Conversations take 20 to 30 minutes and are with program directors, coordinators and GME leaders.
New pilot programs
We are looking for programs willing to run the current version of RESiQ on their evaluation data and tell us what works and what does not. Pilots are free for at least one review cycle.
A pilot covers one milestone evaluation cycle for each resident in a single program. There is no start date yet: for now we are only collecting letters of support while we continue developing the systems. See what a pilot involves.
An earlier version of the tool was used in an Internal Medicine program, and we want to hear from other specialties too.
Best fit: ACGME-accredited residency programs at teaching hospitals and academic medical centers.
About the founder and contact.
Adis Halilovic
Founder, Sabrati LLC
Adis is an MHA candidate and a Project Management Professional (PMP) with years of experience working in healthcare. He is building RESiQ to take administrative work out of resident evaluation review.
Website privacy policy.
Effective October 14, 2026. This policy covers this website, sabrati.com, operated by Sabrati LLC. How RESiQ handles program data is described in How RESiQ handles your data.
What we collect
This website has no accounts, forms, cookies or analytics. If you email us, we receive your email address and the contents of your message. If you book a conversation, the details you enter on the booking page are collected through Google Calendar's appointment scheduling, under Google's privacy policy.
Hosting
The site is hosted by Netlify, which may log technical information such as IP addresses and the pages requested, for security and operations. That is governed by Netlify's privacy policy.
How we use information
We use what you send us to reply, schedule conversations and understand your program's needs. We do not sell your information, and we do not share it with third parties for marketing.
Retention and your choices
We keep emails and booking details for as long as we need them to follow up with you. You can ask us to delete them, or to tell you what we have, by emailing adis@sabrati.com.
Changes
If we change this policy, we will update this page and the effective date above.