We have been using Lyzr for a while now and honestly has been very different than other agent platforms I have tried, the first thing that stands out is how fast it is to get from idea to working agent. No overcomplicated setup, no unnecessary friction. You can actually build, test and run agents without feeling like you need a separate DevOps team. What I like most is that Lyzr is not just about playing around with agents, it is built with enterprises in mind. You can deploy inside your own VPC or infra, you're not locked into someone else's ecosystem (which is a huge problem with other platforms).

They also give you a ton of prebuilt blueprints so you don't always have to start from scratch and it really speeds up experimentation and delivery. Another big plus is how transparent they are. The platform is open at its core, so you know you are not stuck in a blackbox. You can extend it, contribute and actually own your agents.

The team has also been super responsive, has been quick to help out and they actually listen to product feedback. Overall, if you are serious about agents especially in an enterprise setting where speed+control+security matters, Lyzr is a solid bet. It is not just a tool to bet it feels like a long-term asset you can build on. Review collected by and hosted on G2.com.

Right now I feel Lyzr has a very few custom tools available. The base is strong and super agents are powerful but sometimes you want to extend with your own niche tools and library feels a bit limited. I think as the ecosystem grows, this will improve fast. Review collected by and hosted on G2.com.

Lyzr has an easy-to-use and customizable interface, makes excellent AI agents, and has numerous features that allow the product designers and users to make specific agents suitable for their workspace. Review collected by and hosted on G2.com.

It might be complicated for AI beginners or people who are making AI agents for the first time. Review collected by and hosted on G2.com.

Lyzr provides a “low-code / no-code + SDK” environment so companies can build, deploy and manage AI-powered agents and applications without needing to build the entire infrastructure from scratch that is what I felt very amazing. Review collected by and hosted on G2.com.

While Lyzr offers powerful capabilities, that comes with complexity. Organizations without prior experience in AI, data engineering, or devops may struggle to set up and maintain data pipelines, vector-stores or integrations effectively. Like all generative-AI and LLM-based systems, Lyzr powered agents remain vulnerable to hallucinations, i.e. producing plausible-sounding but incorrect or misleading responses Review collected by and hosted on G2.com.

The best thing of Lyzr is its core is an agent framework + agent builder UI (called Lyzr Agent Studio) that lets you build, configure and deploy AI agents (chatbots, data-analysis assistants, knowledge-search bots, RAG-powered apps, workflow automation, etc.) without starting from scratch. Also, t supports both no-code/low-code use (via the Studio UI) good for non-engineers and API/SDK access for developers Review collected by and hosted on G2.com.

Lyzr.ai is very useful and amazing but, it can feel heavy and complex to set up, especially for smaller teams without strong AI or data engineering support. Its agents are still limited by data quality and LLM risks like hallucinations or bias, so human oversight is needed for critical use cases. Integration with legacy systems and ongoing maintenance can increase cost and effort over time. For simple workflows, Lyzr may be overkill compared to lighter automation tools. Review collected by and hosted on G2.com.

Lyzr AI makes customer support easier by handling repetitive tasks, sorting tickets, and giving instant replies. It works fast, provides consistent and accurate help, and lets human agents focus on more complicated issues and building relationships with customers. Review collected by and hosted on G2.com.

Lyzr does have a few drawbacks: it needs clean data to perform well, can struggle with complex or emotional issues, and requires some setup before it runs smoothly. Sometimes its replies may feel a bit generic if not customized properly, and the pricing might be a bit high for smaller teams. Still, for everyday customer support tasks, it’s quite workable and gets the job done. Review collected by and hosted on G2.com.

Lyzr.ai is simple to use and doesn’t require a steep learning curve. I like how quickly it integrates with workflows and the fact that it saves time on repetitive tasks. The interface feels clean, and the automation features make it easier to focus on actual problem-solving instead of manual busywork. Review collected by and hosted on G2.com.

Sometimes it feels a bit limited when you want more advanced options or customization. A few features could be more flexible, and I’ve noticed occasional slow performance. The documentation is also a little light, so if you run into something unusual, you may have to reach out to support. Review collected by and hosted on G2.com.

easy user inteface, not complex to walk around Review collected by and hosted on G2.com.

It is tough to locate created agents and the canvas map with agents goes off after u created it Review collected by and hosted on G2.com.

Lyzr.ai makes it incredibly easy to build and deploy AI workflows without needing deep technical expertise. The platform is lightweight, fast, and offers a clean, intuitive interface that simplifies automation across different use cases. Its modular design lets us integrate AI capabilities into our existing systems quickly, and the implementation time is significantly shorter compared to most other AI tools.

Overall, it’s a very efficient, developer-friendly solution that helps accelerate automation projects and reduce operational effort. Review collected by and hosted on G2.com.

Some integrations still require manual workarounds, and the frequency of updates could be improved to keep pace with rapidly evolving AI needs. Additionally, customer support response times can occasionally be slower than expected during peak hours Review collected by and hosted on G2.com.

The platform's enterprise-ready approach really sets it apart from other AI agent builders. Lyzr comes with built-in Responsible AI and compliance features that ensure data security with full traceability, which gives me confidence when deploying agents in professional environments.

What impressed me most is how quickly I can go from idea to working agent. The low-code approach lets you build and launch AI agents in minutes using pre-built RAG pipelines, eliminating weeks of development work. The platform is also LLM-agnostic and works with any cloud or on-premises setup, so I didn't have to restructure our existing infrastructure. Review collected by and hosted on G2.com.

The documentation and learning resources feel sparse compared to more established platforms. Since it's positioning itself as enterprise-ready, I'd expect more comprehensive guides, tutorials, and case studies to help teams get up to speed quickly. The current onboarding experience could be more robust, especially for teams transitioning from other AI development frameworks.

While the low-code approach is great for quick deployment, I sometimes feel constrained when trying to implement highly customized logic or integrate with legacy systems that don't fit the platform's standard patterns. Review collected by and hosted on G2.com.

As a software developer, what I like best about Lyzr.ai is how fast it lets you go from idea to a working, production-ready AI agent without drowning in boilerplate code.

You get this nice mix of no-code/low-code in their Agent Studio (which is actually great for quickly prototyping or mocking up workflows) and then full SDKs + APIs when you need to go deeper and customize everything programmatically. It's agent-focused right from the start with pre-built RAG pipelines, multi-agent orchestration, guardrails for safety/hallucinations/PII baked in, so you're not spending weeks wiring up LangChain-style chains or fighting prompt engineering edge cases. Review collected by and hosted on G2.com.

As a software developer who's messed around with Lyzr.ai a fair bit, here's the honest stuff that bugs me the most keeping it real and straightforward.

The biggest frustration is the learning curve still feels steeper than it should for something that's sold as "rapid" and low-code friendly. You jump into Agent Studio thinking it'll be drag-and-drop magic, but once you want anything beyond basic prototypes, you end up digging through docs, trial-and-error with components, and tweaking configs anyway. It's not as intuitive as pure no-code tools, and not as flexible/low-boilerplate as raw LangChain or LangGraph when you go full code mode. Feels like it's stuck in this awkward middle ground.

Documentation and examples can be spotty too , some areas are solid, but others leave you guessing how certain integrations or advanced guardrails actually behave in edge cases. I've wasted credits (or time) just figuring out why a tool call or RAG flow isn't doing what I expect. Review collected by and hosted on G2.com.