UiPath Generative AI for Medical Records Summarization

My main use case for UiPath Platform is that we had an engine which had invoice, PO, and GR. We used to get three inputs from the users. We used automation, email automation, Excel, and everything in this application. I'd say this is the best use case.

A specific example of a process I automate with UiPath Platform is whenever we need to check whether the invoice, PO, and GRN are matching or not, as well as a decision engine which is an AI agent that takes the decision and later posts information in SAP as well as sends an email to the user. This workflow works.

The impact of this decision engine and SAP automation on my daily work and team efficiency is that we have automated this using document understanding models for extraction and posting that in SAP. It has turned out to be better results and we have cut down the unnecessary tasks of the humans.

In my opinion, the best features UiPath Platform offers are Excel automation, Outlook or any kind of mail automation. It provides web scraping as well, front end, and it has computer vision as well. It has integration services, API automation, and you can custom build connectors. You have integration services, and you can custom build that. You have document understanding now, you have IXP intelligent process, and you have agentic AI automations. We can build agents, we have UiPath Master, UiPath Flow, case flow, and case. Different kinds of workflows we can build. UiPath has better capabilities for low code and no code as well.

Out of all those features like Excel automation, computer vision, integration services, and low-code/no-code, for day-to-day automations, we use all these things. I would say that the improvements which have happened in document understanding and the agentic, low-code, no-code applications are usually what we use. The improvements which have happened in the document understanding as well as in the agentic side, as well as integration services are standout, and we appreciate them from the UiPath side. We have benefited a lot from these kinds of new features and platform capabilities which UiPath Platform provides for us.

Areas where UiPath Platform can be improved include the licensing part. The licensing at their UiPath end is a complex licensing structure. There is also a steep learning curve which is okay for users who have been using it for a long time, as they understand what it provides and how it can be learned and implemented. If a new user comes in, then it is a difficult task for them to explore all the capabilities. If you want to use all the capabilities with a limited amount of time, it will be a difficult task.

I have been working in my current field for around 3.5 years.

UiPath Platform is stable.

The scalability of UiPath Platform is good. We can automate different kinds of tasks. Every year it comes with some kind of a new feature, and we have a structured licensing as well, with scalable features available.

I would say UiPath has the best customer support, and they have been providing it for a long time.

We have used different solutions, but not for this specific purpose. We have been part of Automation Anywhere and Blue Prism. When it comes to UiPath, it is more user-friendly compared to Blue Prism and Automation Anywhere, and the community and service are also better with UiPath.

UiPath Platform has impacted my organization positively, as we have seen cost savings in the speed at which it delivers with accuracy, and we have got a good ROI as well. The accuracy and the speed and the capability are commendable.

Specific outcomes or metrics I've seen include that we have automated three to five big scale automations. There used to be a lot of manual entries in SAP and these things and extraction as well. We have reduced that to 70% of the manual human interventions. Only 20% remains. The speed at which it works means a low time is required to complete. The chain of work is reduced, so the cost paid on it was reduced. The operations have excelled, the accuracy is good, and it all reflects in the cost savings.

We have seen a return on investment after automating the invoice and PO matching engine. We have reduced it to 80% of employees, and the speed at which it automates, along with the accuracy, brings good ROI as well. We have utilized humans elsewhere, which I would say can be relevant for the position.

My experience with pricing, setup cost, and licensing is that this is one drawback that UiPath has. Previously, there was a better licensing structure. However, when it comes to agentic and new IXP features being launched, there is a little complexity in the pricing and structure that we need to understand. That is one thing I would say to UiPath to improve.

Before choosing UiPath Platform, we evaluated other options such as Automation Anywhere, Blue Prism, and Power Automate.

My advice for those looking into using UiPath Platform is that it provides good platform capabilities right from background automation. You have OCR engines, and you can integrate it with different kinds of third-party services as well. UiPath is better, good, and easy to go. You can start from anything and anywhere, and you have good customer support. The only thing I would mention is the custom setup; other than that, everything else is great with UiPath. I would rate this platform a 9 out of 10.

I have completed multiple projects with UiPath Platform, and the current one I can discuss is Azure Sync, which syncs files present in SharePoint to the Azure container.

For the Azure Sync automation in my daily workflow, SharePoint is maintained by multiple projects and multiple users who upload, delete, or edit files. Since not everyone can access the Azure container and it should be stored as a backup, the bot syncs files every day with two triggers: morning and evening. Whatever users do to the files in SharePoint is synced to the container as well. If a file is edited, modified, or deleted, the same action happens in the Azure storage container to maintain a backup of everything occurring daily.

I can also discuss a project where I worked on KYC verification for credit card appliers with UiPath Platform. I used Document Understanding and APIs to verify images and Aadhaar card numbers, PAN card numbers, and other information.

The agents are now trending with UiPath Platform, and the activities available are numerous compared to other tools I have worked with. I believe UiPath Platform is the best compared to others.

Document Understanding has been helpful in my projects with UiPath Platform.

Document Understanding assists me in KYC verification, where it extracts images, signatures, names, dates of birth, and other information.

UiPath Platform has helped improve our ROI. Most cases have reduced time comparatively.

UiPath Platform helped convert a project's time limits from four hours to 20 minutes, which is a great benefit to the company.

It has reduced human error and freed up employee time. Humans make mistakes that bots will not make, thus reducing errors.

In my current process with UiPath Platform, I experienced an issue with a SharePoint activity and raised a ticket with the UiPath support team. The initiative was good—they came with a call and all—but the issue did not resolve even after two to three months. The support team may need to improve response times and resolution effectiveness.

In documentation, providing examples along with how the activity works would be more helpful.

I have been using UiPath Platform for four years.

UiPath Platform is stable.

It is impressive to see how UiPath Platform's scalability works.

UiPath Platform's customer support is very good. I rate the customer support an eight on a scale of one to ten.

I have been working in RPA for five years. Previously, I was working with Automation Anywhere and now I am working with UiPath Platform.

I evaluated Automation Anywhere before choosing UiPath Platform.

I have seen a return on investment where the bots converted time from four hours to 20 minutes.

The experience with pricing, setup cost, and licensing for UiPath Platform is higher, but it can be justified when compared to other platforms.

I worked with Automation Anywhere and switched to UiPath Platform because users are picking UiPath Platform rather than Automation Anywhere comparatively.

AI agents have been improved regarding UiPath Platform's governance and security.

I have come across UiPath Academy courses and certification courses. Academy is the most valuable resource.

The events and DevCons conducted by UiPath provide significant value to me as part of the user community.

UiPath Platform is easy to learn since it is low-code and no-code. However, it is a vast product with much to learn.

I rate this review a nine out of ten.

I works primarily with Maestro Flow, Agents and IXP to consolidate data from multiple systems during a booking process. Documents are gathered and combined, then IXP extracts the relevant data, which is routed to the Action Center for human review. Once validated, a transformation agent takes over before sending the data to the end system. This blend of humans, agents, and system data substantially cuts manual effort while keeping a human-in-the-loop checkpoint for data validation.

This combination of Maestro, IXP, and agents has enabled projects that would previously have been rejected as infeasible using traditional automation approaches. Many prior proof-of-concepts using conventional methods failed, but this integrated approach has had a positive impact at the organizational level and helped win more projects, despite an initial adjustment period.

IXP (document understanding): Previously, extracting data required extensive training and annotation of documents. With IXP, the process is prompt-based—upload a document, write a clear and precise prompt, then validate and refine. This is especially valuable in scenarios involving 150+ document templates where traditional training-based extraction isn't feasible, since collecting and annotating that many document variants would be impractical. Extraction quality depends heavily on how well the prompt is written, but the time savings over traditional document understanding are substantial.

Maestro (process orchestration): The standout feature is the ability to retry a failed transaction at the point of failure rather than restarting an entire multi-step process. For example, in a 10-process workflow, a failure at step 10 (e.g., an application login issue or maintenance downtime) can be retried directly, without rebuilding the recovery logic that frameworks like REFramework would typically require. Maestro also provides visibility into all running instances and includes livestream demo capability, which has made it easier to showcase workflows to business stakeholders (previously this required screen-sharing from a VM; now it can run unattended with live visibility. Live stream

• Studio 2025 interface: The redesign to mirror Studio Web has made the IDE less user-friendly than previous versions, at least during the initial adjustment period.
• Action Center row duplication: When reviewing multi-column, multi-row data (e.g., 15–20+ columns), users currently must manually re-enter every field for each row even when only one or two values differ between rows. A "duplicate row" feature would let users copy a row and edit only the changed fields, saving significant time. Without it, some users abandon the review and reject records outright.
• Storage bucket timestamps: A feature request has been submitted to expose the write date/time for files in storage buckets.
• Regex visibility in Action Center: Validation conditions written as regex expressions are currently visible to end users in the Action Center UI. Ideally, only a human-readable label should display, not the underlying technical expression. A feature request has been submitted for this as well.

I have been using this solution for four to four and a half years.

IXP had some early model limitations, but recent model versions have performed well with no significant concerns.

Scalability is good. I can leverage all the systems in one ecosystem. All the services in one ecosystem, which is really good.

Support tickets typically result in a direct call from an agent, along with an email trail, ensuring unresolved issues get followed up.

I have not used any other applications. I have gone through the platform of Blue Prism a bit, but I only prefer UiPath Platform and prioritize it as it is a market leader.

Considered straightforward, especially for those working on live projects. Best practices take time to learn, but the core logic is easy to pick up.

We are a partner with UiPath.

Strong ROI, primarily through freeing up FTE capacity previously spent on manual tasks. The combination of agents and IXP enables automation of more complex tasks, including autonomous decision-making, which extends automation beyond conventional use cases.

No info.

When comparing to other platforms, I would suggest 10 out of 10, because it brings the whole ecosystem in UiPath Platform, from building apps and agents and LLMs, everything is available in one platform, including the orchestration.

UiPath Platform brings the whole ecosystem in a single platform, from the Studio to orchestration, from the apps and the dashboards and also Maestro, extraction, LLMs, everything is coming under one platform. Now I can do some basic apps or dashboards within UiPath Platform, which is really helpful. Overall, when I see this ecosystem, it is all interconnected and it is useful. UiPath Platform is very much user-friendly when all the services are coming under the orchestrator or the Studio Web or Studio and it is all interconnected. I do not have to go for any other platform to get all those. It is really an amazing solution.

• UiPath Academy is considered a mandatory, high-value resource for staying current on both new and legacy features, offering practical, example-driven content useful for beginners and experienced developers alike. That said, hands-on project experience is viewed as more valuable than academy content alone.
• UiPath community (workshops, meetups) is seen as a major differentiator, helping users stay current and experiment with new features.
• Agent usage is still limited—so far used mainly for a simple name-mapping/conversion task via database lookup. Proof-of-concept work with agents has gone well.
• Agent trust layer / data trust layer: Data security is good, with masked data feeding into LLMs ensuring strong data privacy.