UiPath Generative AI for Medical Records Summarization

My main use case for UiPath Platform  is invoice processing. An agent helps identify relevant inbound emails/attachments (invoices), then a robot validates and posts invoices into two or more Saas or app (i e:ERP  platforms). Document Understanding/IDP  extracts key fields (vendor, invoice number/date, totals, tax, line items) before human validation and final posting to finance.

UiPath helps by providing an enterprise-grade framework for governance and auditability in regulated contexts. It reduces the effort of rebuilding controls for each automation, supports standardized delivery, and improves reliability and consistency when handling client-facing processes.

UiPath is a leading platform for enterprise automation. For my use cases, the main improvement area is agentic/LLM-based automation maturity in production; especially consistency, controllability, and reducing the iteration required to reach stable results. Continued enhancements to guardrails and human-in-the-loop workflows would further strengthen large-scale deployments.

I have been using UiPath Platform  for seven years.

UiPath Platform is the best in terms of stability.

Regarding UiPath Platform's scalability, I have never tried scalability with a hundred or thousand robots, but they say it is easily scalable. I have never had those clients where I could put a hundred to a thousand robots, but they say it is scalable easily. I trust them.

The customer support for UiPath Platform is good.

Before choosing UiPath Platform, I evaluated options such as Blue Prism  and Automation Anywhere .

I have seen a return on investment of around a huge return in three to six months, which is a short time regarding the usual SaaS products from two to three years ago or 18 months prior. Now with UiPath Platform, you can deliver a return on investment in three to six months. It depends on the complexity of the workflows.

Overall, the experience with pricing, setup cost, and licensing has been acceptable. The key is to size the licenses to the automation strategy and expected utilization to ensure predictable ROI.

I evaluated Blue Prism  and Automation Anywhere , but selected UiPath for its overall ecosystem and delivery speed for enterprise automation—particularly around development experience, orchestration, and document processing for end-to-end workflows.

I would rate UiPath Platform 9/10. Start with a high-volume, well-defined process (e.g., invoices), define ROI metrics upfront (cycle time, volume, error rate, cost of errors, build/maintenance), and scale gradually with clear governance and reusable components.

My use case involves developing robots because I worked as a consultant before and worked with multiple domains like banking, insurance, and healthcare, where I automated monotonous processes. This is my use case where we build a robot, and that is what is mainly used for this tool.

What I appreciate about UiPath Platform  is that it is very user-friendly and continuously updates the product according to market trends, which I prefer compared to other competitive tools.

UiPath Platform  has reduced human error and freed up time in many cases where we have removed the entire department, and in many cases we have freed half of their work. As a consultant, I can explain this in technical or non-technical terms. We do not talk about the hours we saved; we talk about the FTEs, which means the full-time employment we saved. I think we save approximately five or six FTEs in a year.

An area where UiPath Platform has room for improvement is that they do not have a local source code saving opportunity where we need to save it in our local file or a local PC. They do not have that capability compared to Blue Prism , which directly saves the source code in a database. In UiPath, they don't have that opportunity. Even if I lose my laptop and develop personally, if I lose my laptop, then the entire source code is gone. The tool should not be device-dependent; it should be a cloud-dependent tool.

I have been using UiPath Platform since 2017.

My thoughts on the stability of UiPath Platform are that sometimes we have an issue where they do an update from their side that impacts our robots, especially on a weekend, which is a disaster. I would give it a 4.5 out of 10.

I would rate the scalability of UiPath Platform as highly scalable because they have already integrated a lot of AI features inside the tool, which is truly appreciable. On a scale of 1 to 10, I would say 9.

I would rate their technical support as an 8 out of 10.

I would recommend UiPath Platform to other users, and the first important advice I would give is to check out UiPath Academy, which is very good. Also, the community edition, which is available online from UiPath, allows us to use the tool and learn on one single platform. That is only provided by UiPath. I recommend UiPath Platform because it is user-friendly and cost-effective, and we can get our ROI within six months.

I have not started using the agentic automation in UiPath Platform because I took a course last year at a conference, and I think we have not got any use case for that. I am in dialogue with my management on how we use that going forward because it is still emerging, so we have not got any use case for that.

My thoughts on UiPath Academy courses are that the first thing is they are free, and the second thing is they have a lot of useful insights where it gives the complete product overview and what is included. The missing part is that as a course, we need proper understanding of English or proper communication skills, which is boring when compared to all the other courses available online.

When I compare UiPath Platform with other products, I see that everything is perfect with UiPath Platform in terms of pricing, user-friendliness, capability, and efficiency. I would give it a 9 out of 10, while for other products, I would give 7 or 6 because there are a lot of other flaws in those tools. The only problem I have is that it is device-dependent, and that is the reason I have taken one point down from UiPath Platform.

On a scale of 1 to 10 for pricing, I rate UiPath Platform between 6 to 7. UiPath made a complex design on the licensing part. Each license has different pricing, such as a developer license, production license, or tester license. They made it not very simple. If you compare with other tools, they made it very simple. You buy a production license and get a developer license along with that for free. I think that looks very simple.

The hands-on users of UiPath Platform in my organization would be around 15 to 20. It is a public sector in the whole of Denmark. If I include that, we have around 300 or more developers in the complete Denmark public sector, and in my department, we are around 10. For usage, we have around 300 or more robots.

I do not think the solution requires maintenance; it is not a complex one. UiPath made it very simple, and I appreciate that.

My take on the AI aspect of UiPath Platform is that it is the next future. I think it is also an emerging area where people think AI can do everything, but it is not true; it still needs to be checked. The robots are the main important thing for sourcing the data to the AI. Both are very important. AI is the future, but not so early.

I would rate this review an 8 out of 10 overall.

My use case is automating all manual processes that our clients face difficulty working on, such as data extraction and automation of all data entries. I automate all kinds of manual efforts for the clients.

From my experience, UiPath Platform  has reduced human error by removing manual work, and it has reduced almost 70 to 80% of manual efforts. One of our clients who works for Tally used to spend the entire day completing ledger solutions, but after automating that, it was completed within one to two hours, saving almost 70 to 80% of the person's time.

I use agentic automation to help with challenges, and I also use agentic AI in UiPath Platform . In my opinion, it is a good thing because when we use agentic AI, the AI-powered agent observes, decides, plans, and acts automatically to complete tasks instead of following fixed-based rules for the workflows. I appreciate the agentic AI workflows.

With agentic automation, it helps to leverage people, robots, and AI agents to automate processes by having Document Understanding integrated with LLMs, such as Action Centers. I value the LLM parts for that.

The best feature in UiPath Platform, according to me, is the intelligent automation platform that combines modern RPA , AI, Orchestrator, and a strong governance ecosystem in one. We are using modern UI activated, unified targets, and it acts as a single-click activity type, where Orchestrator serves as the central control of all the robots, scheduling, and monitoring, while also having assets for securing our credentials. Additionally, it includes VPN work, a built-in retry mechanism, structured explanation management, easy scaling of queues, Document Understanding for invoice handling and PDFs, ML-based extractions, agentic AI, and autopilots for natural language automations and fewer rule dependencies. That is why I feel UiPath Platform is the best.

The improvements I have noticed in UiPath Platform over the last three years are significant. When we started using UiPath Platform, it was on the classic mode, and now it integrates with modern solutions, offering a UI and AI-native development experience with features such as AI Builder, screenplay, APIs, workflows, and UX enhancements. They are working on fresh designs, updating panel docking, and navigating through logical and sequence views. I see stronger automation on cloud capabilities with automatic cloud continuous updates, focusing on indexing, contacting, and AI automation. The new addition of agentic AI is also invaluable for agenting apps and agentic extractions. While Orchestrator and other features were already present, they are now significantly upgraded, including cloud testing and agentic testing, which reflect the changes over time within three years.

In terms of areas for improvement, UiPath Platform needs to address the high licensing costs in India, making it difficult for small teams to estimate and manage the expenses. Additionally, there is room for better performance in UI automation as it sometimes feels slow for heavy web applications. Upgrading browsers often breaks selectors, so UI performance and the browser dependency can be optimized further. We should reduce reliance on fragile UI clicks for UI automation. I believe we need a stronger API-first integration approach and increased native connectors as many tools such as Studio, Data Understanding, AI Center, and Orchestrator can overwhelm new users. We can provide learning paths inside Studio for smart automation templates and improve the stability of desktop automation for legacy and Citrix automation. Improvements in OCR accuracy and image recognition speed are also necessary. Furthermore, we can enhance transparency in agentic AI decision-making to avoid "black box" scenarios and boost confidence in scoring visibility. While UiPath Platform has no major weaknesses, areas for simplicity and transparency remain to be improved.

I have been using the solution for the last 2.9 years.

Regarding stability, I rate it at 10. UiPath Platform is very stable without significant issues.

It is scalable for my business, but I provide a rating of nine for the scalability aspect.

The user community is a better place for addressing issues where we can post our queries, and the community is always helpful in providing solutions. That is a good aspect.

I have used UiPath Platform, Power Automate , Blue Prism , and a little of Automation Anywhere . In my opinion, for any mid-scale to large enterprise solution, UiPath Platform is the best among these options and remains at the top of the competition. However, I feel that Power Automate  has a good advantage for cloud-to-desktop and desktop-to-cloud automation.

The deployment process itself is easy and not too complicated. For someone who has used UiPath Platform before, it is straightforward. However, for beginners, it can seem quite challenging at first, but eventually, it becomes manageable after a few deployments.

There are mainly 10 to 12 persons using UiPath Platform.

Regarding the Academy courses, I gain a lot of new knowledge from Academy automation. I am updated with all the latest information, gaining insights into how to utilize UiPath Platform effectively, which provides additional learning resources for us. Academy is a great asset as it allows users to stay well-informed about new updates without stress.

I would recommend UiPath Platform to other users because it is proficient in AI, Document Understanding, and a range of other functions, making it a valuable tool. However, I also recognize that many clients in India face budget constraints regarding such expensive solutions, indicating that UiPath Platform needs to consider pricing for wider adoption. I have provided an overall review rating of 8 for UiPath Platform.

Today I faced an issue installing the unattended robot in the production machine. I signed in with the robot account and entered the client ID and secret. The system did not recognize the license and was not giving me a proper response. I did not understand what the actual issue was and searched for the community guidelines as well. I had no issues when configuring a business user, an automation developer, or any other users, but I felt discomfort when onboarding a robot account into UiPath Platform  Orchestration and connecting that with the production machine.

I am not using Agentic automation in any of my production environments, but I am using it for development purposes. I am exploring the platform on the Agentic side. I was working on a use case where I wanted to extract meaningful data from a chunk of error messages. I used Maestro for that and it was working very well, so the user can easily correlate with the actual issue.

The Agentic response when using some ML activities or AI activities is getting generated with Maestro. Those are up to the standard. I used Maestro to extract meaningful data from error messages and it worked very well.

The developer-friendly environment and the resolution to the problems faced inside UiPath Platform  are very fast, and the customer support service as well as the development-friendly environment make deployment and development processes easier.

One thing which I can suggest is to integrate the CI/CD pipeline within UiPath Platform Orchestration. That will be the next step to get into a full Agentic service within UiPath Platform. I am currently using some external CI/CD pipelines, like Azure  or others, and integrating them inside UiPath Platform would be an easy-to-go solution for anyone.

The sales approach is a bit slow. It almost took a month to complete the entire sales process and just get the final quote from UiPath Platform US sales team. There was also some discomfort in selecting the packages. The AI unit bundle flexibility in pricing is missing. I wanted to buy a smaller portion of AI units but needed to purchase a large bundle.

Recently, I faced an outage issue with Cloudflare , resulting in being unable to connect UiPath Platform with UiPath Platform Orchestration, and some of the UiPath Platform Orchestration services were down. This was the first time I faced a stability issue.

I expected faster support during a production deployment issue. I got email communication, but a call would have been faster, resolving the issue more quickly.

In the past six years, I never faced an issue, but in the last couple of months, I faced a Cloudflare  outage issue where I was not able to connect UiPath Platform with UiPath Platform Orchestration.

In the last couple of months, I faced a Cloudflare outage issue leading to problems connecting UiPath Platform with UiPath Platform Orchestration, along with some services being down. This was my first encounter with a stability issue after six years of experience without any outages.

Scalability is not an issue. Configuring the machines is straightforward with the use of machine templates. After the first configuration, scaling the same process to multiple machines just involves assigning users and adding the machines to the templates, making it an easy process.

I used to connect with technical support. Recently, I faced an issue in configuring the templates and assigning the user from the UiPath Platform Orchestrator tenant. I contacted the technical support team, and they assisted me in resolving that issue quickly through email.

The reason for rating eight is that the email support was good but slower than expected when I had a production deployment issue. I expected quicker response times, preferably through calls, which could resolve issues faster.

I used AssistEdge, an RPA  tool from Infosys, and Power Automate , which is from Microsoft.

The initial setup, including the machine setup, takes at least a day, but if the machine is already set up, it is just a one to two-hour task.

Only one person will do the deployment.

Pricing is a bit costlier compared to Power Automate , but the same as AssistEdge. It delivers as expected, so the charge is acceptable.

For some cases, I prefer Microsoft Power Automate because of its cloud-integrated connections, such as creating a flow triggered by a Microsoft email, which is not as straightforward in UiPath Platform. Integration services could be improved within UiPath Platform.

Deployment is a straightforward task, but ensuring integration of the CI/CD pipeline within UiPath Platform Orchestration would enhance the ease of use. I rate this review an eight overall.