I mainly use UiPath Agentic Automation for creating agents that handle repetitive tasks and support different business processes, which helps reduce manual work and makes automation more flexible and efficient. I love how it combines AI with automation, simplifying the creation and management of agents, and it helps automate complex tasks with less manual effort. The AI integration makes automation more adaptable as it can handle tasks that aren't always strictly rule-based and reduces the amount of manual configuration needed. Creating and managing agents is straightforward, saving time and making it easier for my team to build and scale automation across different processes. Plus, it works well with other business tools and applications, connecting workflows smoothly across multiple platforms. Review collected by and hosted on G2.com.

Sometimes the setup and configuration can feel a bit complex, especially for a new user. A few features also take time to understand properly. I think the platform could improve the learning curve with better documentation and a more simplified user experience in some areas. Review collected by and hosted on G2.com.

What I like most about UiPath Agentic Automation is how it brings AI, RPA, and human-in-the-loop processes together in one unified platform, enabling true end-to-end automation.

The drag-and-drop workflow designer and low-code approach have noticeably reduced development time and made it much easier to collaborate with non-technical users. Features like Document Understanding and AI Center help automate unstructured work such as invoices and emails, and the agentic capabilities let bots make smarter, more context-aware decisions with fewer exceptions.

UiPath also integrates smoothly with enterprise systems like SAP and with APIs, so cross-platform automation feels seamless without requiring heavy customization. In our experience, the platform has been stable and scalable, and it has delivered clear time savings and ROI within a short period.

On top of that, UiPath Academy and the community support made onboarding straightforward. We were able to move quickly from basic automation to more intelligent workflows, and the business impact was stronger than we expected. Review collected by and hosted on G2.com.

One challenge with UiPath Agentic Automation is the complexity of setting up advanced AI-driven workflows, which often requires specialized expertise and can slow initial adoption.

Debugging multi-agent workflows could also be more user-friendly, as tracing issues across bots and integrations isn’t always straightforward. We’ve noticed some latency in highly complex, integration-heavy processes, and the pricing model may be limiting for smaller teams looking to scale.

Improving usability, debugging visibility, performance, and pricing flexibility would further enhance the overall experience. Review collected by and hosted on G2.com.

The new Temporal as a Service that helps users to create Multi agent orchestration Review collected by and hosted on G2.com.

Nothing as such all the new features are good but the documentations are not that ellaborative for new features Review collected by and hosted on G2.com.

One of the newer things I like about UiPath Agentic Automation is how the platform is evolving from simple automation into a fully orchestrated, multi-agent system.

Recent updates like the Maestro orchestration layer make it much easier to coordinate multiple AI agents, bots, and human steps in a single workflow, improving visibility and control across complex processes.

I also like the introduction of Agent Builder, API workflows, and unified development in Studio, which simplify building and scaling agentic solutions without switching tools.

Another standout is the progress in intelligent document processing (IXP)—with agentic extraction, validation agents, and autopilot features that can generate schemas and improve accuracy over time, making document-heavy workflows much faster to implement.

Additionally, newer capabilities like natural language-driven automation (e.g., describing tasks to build workflows) and agentic testing/self-healing automation are making development more intuitive and reducing maintenance effort.

Overall, these updates make the platform feel more like an intelligent automation ecosystem, where agents can collaborate, adapt, and scale—rather than just execute predefined tasks. Review collected by and hosted on G2.com.

Pricing and licensing flexibility is area for improvement—many users find it expensive, particularly for smaller teams or gradual scaling. Review collected by and hosted on G2.com.

I like how UiPath Agentic Automation combines AI, RPA, orchestration, and human collaboration into a unified automation ecosystem. It enables workflows to make contextual decisions, adapt dynamically, and handle more complex enterprise scenarios. I appreciate the orchestration capability using UiPath Maestro and AI agents, which allows multiple intelligent components to work together efficiently, making automation solutions more autonomous and scalable. I also like that it supports Human-in-the-Loop validation, which balances AI-driven automation with controlled human oversight, making it practical and reliable for enterprise environments. Another aspect I value is its flexibility in handling semi-structured and unstructured data intelligently. The ability to automatically generate structured documentation like PDDs and SDDs reduces manual effort and speeds up project delivery. Overall, I find that UiPath Agentic Automation enables creating intelligent digital workers capable of reasoning, collaborating, learning from context, and supporting decision-making beyond merely executing predefined tasks. Review collected by and hosted on G2.com.

While I see strong potential in UiPath Agentic Automation, there are still a few areas that could be improved based on my hands-on experience with AI-driven exception monitoring and intelligent PDD/SDD generation use cases. One challenge is the predictability and consistency of AI-driven decisions. Since agent responses can vary depending on prompts, context, and input quality, maintaining deterministic behavior for enterprise-grade automation can sometimes be difficult compared to traditional rule-based workflows. Another area for improvement is debugging and observability. In complex workflows involving multiple agents, orchestration layers, prompts, and human validation stages, tracing why a specific decision was taken can become challenging. More detailed reasoning visibility and monitoring would help significantly in production support scenarios. In my intelligent PDD and SDD generation use case, one practical limitation I observed was that the agent could generate structured content and process understanding effectively, but it was not able to directly generate properly formatted DOCX documents with enterprise template formatting requirements. To address this, I integrated a traditional RPA bot into the workflow. The AI agent generated the required structured data and document content, while the RPA bot mapped those values into predefined PDD and SDD Word templates by replacing placeholders dynamically. This hybrid approach worked effectively, but stronger native document generation and formatting capabilities within the agentic framework would further improve end-to-end automation. I also feel prompt lifecycle management and reusable accelerators can be enhanced for large-scale enterprise implementations. As agentic solutions grow, maintaining prompts, validating outputs, and governing AI behavior across environments becomes increasingly important. Review collected by and hosted on G2.com.

I find the governance of code with UiPath Agentic Automation's multi-agent system architecture great, as it makes API scripts flow like neurons, enhancing decision-making and observability. The lightweight AI agents are a plus, structuring unstructured data and handling unexpected system exceptions effectively. It also excels in handling unstructured data with ambiguity and complex decision-making, which are areas where traditional RPA solutions fall short. Review collected by and hosted on G2.com.

Orchestration is a solid point of the whole UiPath environment, but where developers need custom controls, it needs improvement as developers have to work on highly customized toolkits and languages all the time. JSON schema validation is something we need to do better for customization controls. The Python runtime should be improved; while I understand UiPath is heavily built on the .NET technology stack, Python is a highly adopted language by my clients/customers, so I need its response to be more quick. Review collected by and hosted on G2.com.

I use UiPath Agentic Automation for intelligent automation and orchestration, and I really like how it minimizes coding efforts with its 'just upload PDD and done' approach. The easy and good UX makes the experience better. I appreciate its ability to seamlessly combine autonomous, context-aware decision-making with the robustness of traditional RPA. The Proactive Goal-Oriented Autonomy is impressive as agents can interpret context, reason through problems, and adapt to changing conditions in real-time without constant human oversight. It manages end-to-end processes with orchestration of human-in-the-loop, involving people only when needed, which enhances productivity in complex scenarios. Additionally, the secure and governed framework, built for enterprise adoption, integrates well with the existing UiPath infrastructure to maintain auditability, compliance, and control over AI agents. The setup was easy too. Review collected by and hosted on G2.com.

Cost Review collected by and hosted on G2.com.

I like how UiPath Agentic Automation's workflows operate with natural language prompts, which means there's no need for deep technical knowledge. It reduces the human intervals for automation, providing strong automation workflows and making success achievable with less effort. I've used the natural prompts in two categories: during test cases when I'm stuck with conditions or need to verify activities, and for designing custom agents. The biggest help I've received from UiPath Agentic Automation is that simply by providing a natural language prompt, I'm able to create entire workflows with successful results. The Auto Pilot feature in the Test Manager of Test Cloud is another aspect I enjoy, as it assists me in creating requirements, test cases, and analyzing them, along with using suggestions. Initially, it took time to understand everything, but now it's easy to use, and as with learning anything new, it just takes time. Review collected by and hosted on G2.com.

Initially it took time to understand all things, indicating a learning curve. Review collected by and hosted on G2.com.

I use UiPath Agentic Automation to build multi-agent orchestration workflow activities, and I find the orchestration maintenance and memory management to be very well handled. This helps me write less code and focus more on functionality and execution. I think the new Temporal as a Service that UiPath recently launched is awesome. It features a replay mechanism of LLMs and event storage, allowing you to start from where the flow stopped without needing to start over. This is particularly useful when there's a HiTL delay, as it keeps track of events and helps lower the cost of tokens considered by LLM. Also, the initial setup was super easy. Review collected by and hosted on G2.com.

The documentation can be improved by adding more use cases. By adding more blueprinting solutions, it would help users to start building those. Review collected by and hosted on G2.com.

I really like the simplicity and ease of use of UiPath Agentic Automation. The low-code customization allows users to configure smart agents quickly without the need for complex coding. This democratizes development by enabling business users to deploy autonomous bots independently, reducing their reliance on specialized IT departments. I also appreciate the natural language control, where admins can instruct digital workers using simple multi-step prompts instead of fragile logic loops. Additionally, I find its extensive integration with a broad ecosystem of enterprise software, third-party AI frameworks, and LLMs to be quite beneficial. Review collected by and hosted on G2.com.

Runtime Uncertainty and Governance Pain. Non-Deterministic Failures: Shifting away from strict rules means AI agents introduce runtime uncertainty. They occasionally produce unpredictable outputs or fall victim to Large Language Model (LLM) hallucinations. Debugging Complexity: Traditional RPA workflows have transparent, step-by-step logs. In contrast, tracing UiPath Autopilot decisions or agentic actions can feel like a 'black box'. This lacks clear explainability for auditing. Intermittent UI Failures: Agents interacting with dynamic user interfaces via computer vision occasionally hit timing or selector mismatches, requiring developers to manually wrap logic in heavy error-handling or retry scopes. Review collected by and hosted on G2.com.