WebRobot’s Leadership Path

This is the first of a series of articles to tell you about the current state of our project.

Note: this post describes our early roadmap. An update on where that path has taken us follows at the end.

Those who already follow us through social networks know that we are close to the first release of our managed service, thanks to which we will be able to offer periodic data acquisition activities on subscription. But in this very particular section, our goal is to show you the path we want to achieve to become the leaders in this specific sector, taking advantage of strategic partnerships with big-data companies, universities and research centres.

While the lite version of our service will allow you to define deterministic query queries, our research and development efforts will involve machine learning algorithms and automatic extraction techniques aimed at overcoming the limitations of the current state of the art.

We want to take advantage of our connection with the CNR and researchers who are experts in Wrapper Induction techniques, to go where our competitors have not yet seen each other.

The automatic induction of interactive actions is in fact our greatest ambition, so much so that this algorithm will inevitably be protected by a patent for our exclusive commercial use.

The creation of automatic induction algorithms will allow the scalability of our service to be maximized, which already on the hardware / infrastructural level can rely on the robustness of the AWS ecosystem.

Stay tuned and follow this specific channel, there will be plenty of reasons to keep your neural circuits buzzing.

WebRobot Ltd wants to be a point of reference for that community of data scientists, attracted by this fascinating world, from which unexpected, creative multidisciplinary connections can emerge.

Despite the patent that we would like to implement, the algorithms that will emerge from this path will be associated with specific scientific publications that we will release in this channel without hesitation.

Come back later for our next post. See you soon!

Update: where the path led

The ambition of inducing extraction rules and interactive actions automatically has since become the core of the product. Today WebRobot is a console where AI agents build data pipelines from a plain-language description and validate them on real rows before they run, with scheduled pipelines for periodic acquisition. Agents rely on a self-healing acquisition catalogue of 610 sources across 34 markets, collected with real browsers and within each source’s terms of use.

On the infrastructure side, the platform can run in the customer’s own cloud (BYOC), and multi-agent teams run on Ray with human approval on critical steps. Read more about the platform and the verticals built on it.

Request your invitation