Table of contents
Waves of change
Many people think and speak of AI as one big development, but the reality is more nuanced. AI has been around for decades, however, the specific technologies that disrupt our world seem to be coming in waves—let it be generative AI, agentic AI, or whatever comes next.
Those new technologies can be huge opportunities, if you can leverage them. To stick with the wave analogy: You can either catch and surf those waves, or they crash on top of you and wipe you out.
This prompts many executives to spring into action, being concerned that being passive means taking the risk of missing out. However, AI technologies are incredibly powerful tools, and like with any tool, you can both create and destroy value in the blink of an eye. It is not just about taking action, but taking the right action (and avoiding the wrong ones).
But how can you take the right action with such rapid developments, when it’s hard to predict which specific AI wave might come next?
The answer is to create an operational, or process, architecture, that allows for two things:
Firstly, to realize value today, so that you can incorporate and benefit from new AI technologies when they are actually ready to drive value (Gartner refers to this stage as the plateau of productivity).
Secondly, to be ready for tomorrow, which means your process architecture is adaptable and can incorporate the “next thing” even if we don’t know it yet. This requires a future-proof degree of flexibility, scalability, and resilience.
Now, how can you put in place such a process architecture, when you’re facing the “spaghetti architecture” of historically grown IT system landscapes?
Well, the wrong answer would be to “just throw in AI.” As silly as it sounds, that’s what we’re often observing when executives spring into action out of fear of missing out (FOMO). In fairness, it can lead to some short-term value gains, similar to what we have seen with Robotic Process Automation (RPA) in the past decade. But just like with RPA, it also leads to isolated point solutions, which are hard to maintain, change and evolve, and can eventually stall the organization—a value trap. In other words, “throwing in AI” will create the next layer of technical debt and make the problem worse.
This is where process orchestration comes in. A process orchestration layer hovers above all involved people, systems, and devices. It controls the flow of the end-to-end business process and ensures that everyone does their part, even when unforeseen incidents occur.
Process orchestration doesn’t replace point solutions, but integrates them. It allows you to tame the complexity of your system landscape and overcome the “spaghetti architecture,” align business and IT, and ensure an overall flexibility to adopt new technologies.
Analysts like Forrester confirm that process orchestration is needed to adapt quickly and innovate:
Gartner even sees a new category of software products emerging, which they call “Business Orchestration and Automation Technology (BOAT),” with process orchestration as the very key differentiator at its core.
What to expect from Camunda
The foundation: Unrivaled process orchestration
It’s clear that process orchestration is the key enabler to operationalize AI. It has always been the focus of our product innovation, and will continue to be. We’ve created the by far most capable process orchestration engine in the market, to provide you with the flexibility, scalability, and resilience that you need in the age of AI. It supports not just simple workflows, but complex, long-running end-to-end processes (which, depending on the use case, is sometimes also referred to as case management). Thanks to the open BPMN standard, we bring business and IT together, and our customers can adopt Camunda on-premise or in their private cloud as well as our managed service.
Another key differentiator of Camunda is composability: From day one, we created our platform based on the principles of loose coupling, for example by inventing the job workers paradigm, which provides flexibility and extensibility. Thanks to this approach, it is very straightforward to combine our process orchestration engine with any automation technology on the task level, and we have done so in previous years, by adding:
- Human in the Loop
- Decision automation with DMN
- Integration capabilities for both microservices, and (low-code) Connectors for APIs, including RPA products.
We will continue to invest in those, but we’re also expanding the coverage of our platform, as you will see below.
Adding native support for RPA and IDP
Thanks to our composable architecture, our customers can combine Camunda with anything, and most of them take advantage of that. Over the years, we have recognized usage patterns that inform us which technologies are most obviously a complement to process orchestration. In the past 18 months, Robotic Process Automation (RPA) and Intelligent Document Processing (IDP) were increasingly leading the pack.
Let’s double click on RPA, since the underlying principle applies to IDP just as well.
Many of our customers combine Camunda with RPA products to tame the complexity of siloed automation, and they approached us with the following statement: “We don’t need a ‘full’ RPA product. We just need a simple way to integrate legacy applications that don’t expose an API.”
In fact, some of them actually made the effort to strip down their RPA bots, to “move the intelligence from the task (RPA) layer to the process (Camunda) layer,” and in a few cases even created their own, simplified version of an RPA product. Once we saw that pattern emerge, we saw the clear opportunity to provide a solution out of the box.
That’s why I’m happy to announce the following new capabilities of Camunda:
- Camunda RPA (Robotic Process Automation) – Transform isolated task automation into comprehensive end-to-end process orchestration by embedding RPA directly into a workflow. Camunda’s composable architecture seamlessly integrates both existing and newly developed bots within its RPA services, all within the same platform.
- Camunda IDP (Intelligent Document Processing) – Replicate human work using AI and ML to classify, extract, and process information from documents, minimizing errors and saving valuable time. Organizations can easily integrate IDP into their processes to gain full visibility into their end-to-end business processes.
Both are currently available in version 0.1, so you can evaluate them while we’re completing version 1.0 as part of Camunda 8.8 (scheduled for April 2025). Both are being realized following these core principles:
- Lightweight: They contain the essential core functionality, no fluff or complexity.
- Easy to use: That makes them an easy to use solution for most use cases.
- Pre-Integrated: They’re part of the Camunda platform and available out-of-the-box.
- Observable: You can monitor both RPA and IDP tasks as part of the end-to-end process.
- Composable: You can use Camunda RPA / Camunda IDP, but you don’t have to. We keep maintaining our powerful Connectors for the leading products in the market (and might add more), and allow you to connect with your own homegrown solutions leveraging the Connectors SDK.
As excited as I am about these new possibilities, I need to say that they are actually just the beginning, as we’re expecting to add more capabilities to the task automation layer in the months and years to come. Stay tuned!
If this sounds interesting, please get in touch with us and we’re happy to discuss your requirements and give you a live demo.
Business solutions provided by Camunda and partners
With the right foundation (process orchestration) and building bricks (task automation capabilities) in place, we can drive value even faster with solutions that solve business problems right away. Those are provided by both partners and ourselves. Here are just a few examples:
On the Camunda Marketplace you will already find close to 100 connectors to integrate various applications, from OpenAI to Salesforce. You will also find a rapidly growing pool of solution blueprints, which are informed by real-world projects and give you a way to kick-start the development of your own solutions. Some provide reusable process patterns, for example AI-powered routing of customer enquiries. Others cover complete business process solutions, including BPMN process models, DMN decision tables, and task forms, like this Chatbot-driven loan origination process.
For our customers using SAP, we have a particularly interesting new solution: Process Orchestration for SAP integrates S/4HANA and ECC into Camunda processes, so that you can call SAP BTP (Business Technology Platform) Services from Camunda processes and render Camunda task forms in SAP Fiori design in SAP BTP, which means your SAP users are involved in end-to-end processes orchestrated by Camunda while staying in their usual SAP environment.
This allows you to standardize your processes in BPMN and DMN, no matter if they run inside or outside of SAP. You increase your flexibility while keeping the SAP Core “clean,” by empowering your business to own and change the business process and decision logic.
Based on customer input, we’re seeing this as a particularly valuable solution for organizations that are struggling to migrate to S/4 HANA, because it allows you to keep your processes stable, while modernizing the underlying SAP infrastructure.
If this sounds interesting, please also get in touch with us and we’re happy to discuss your requirements and give you a live demo.
Embedded intelligence
We’ve already established that a process orchestration-based architecture is the right approach to operationalize AI. However, our vision for AI goes much further.
We’re seeing AI not as a component, but a principle that permeates our entire platform, i.e. it’s going to be present in every part of it. This can be explained along three areas of AI-specific investments that we’re currently making.
- Fast-Track to orchestrated processes, to help you create running process applications much faster. One key ingredient here is the Camunda Co-Pilot.
- Easily integrate AI in your existing process by leveraging the Connectors ecosystem as well as our new components, for example for Intelligent Document Processing.
- Intelligent execution and optimization, which eventually leads to autonomous orchestration. The big question will be: How can we benefit from the ad hoc nature of agentic AI, without taking the risks of unpredictable behavior? To answer that question, we are actively exploring how you can leverage the BPMN standard to blend pre-determined with ad hoc, i.e., agentic AI-driven process orchestration. We’re calling this autonomous process orchestration with guardrails.
Again, if this sounds interesting, please get in touch with us, to get a live demo of the AI capabilities that are already there, and to discuss what you can expect from us in 2025 and beyond.
Summary
The AI waves keep coming, and they’re hard to predict. With the right process architecture, you can realize value today, and be ready for whatever comes next. This will be your superpower to operationalize AI.
Camunda enables you with a composable Process Orchestration and Automation Platform with embedded intelligence.
Our foundation is and will remain process orchestration, and we are expanding our platform to all aspects of automation. Camunda’s key differentiators are:
- Composability with an integrated yet flexible platform.
- Embedded intelligence to fast-track development, let you orchestrate and integrate any AI technology, and realize a harnessed, i.e. reliable and secure, autonomous orchestration.
- Open standards like BPMN and DMN to support Business-IT collaboration with one model, designed in a shared language.
- An execution engine that is horizontally scalable, cloud-native, and highly resilient, hence battle proven for your mission-critical core processes
I do hope that you’re as excited about the future as we are 🙂
