Robots serving dishes in Karachi: what they really do

In Karachi, Robia and Jojo bring dishes to the tables, but taking orders and clearing plates remain human tasks. A concrete example of service robotics, distinct from functions announced for the future.

In Karachi, two robots designed in Pakistan bring dishes to the tables of a restaurant. Their role is practical but limited: they carry the plates, while customers or staff take them off the tray. The scene offers a glimpse of service robotics without suggesting that the dining room is fully automated.

Two machines among the tables

Robia and Jojo move between the tables at The Sicilian restaurant in the Pakistani metropolis of Karachi. Robia introduces itself with an electronic voice; the two machines can therefore help welcome customers and bring dishes to them. But the final step remains a human one: customers must take the food from the tray themselves or wait for a member of staff to help.

This distinction helps clarify what a service robot actually does. It can move an object through a busy space without taking an order, preparing a meal or managing the customer relationship on its own. In the case of Robia and Jojo, the available information describes mainly the delivery of dishes and a simple introduction to the restaurant. It does not demonstrate general autonomy in dining-room work.

The robots were developed by Pakistani engineer Syed Osama Aziz, who also founded the restaurant and robotics company Cobot. According to the AFP report published by RTL Today, the machines were developed using mainly locally manufactured parts, imported electronic components and circuit boards designed by Aziz. He says that some of the equipment needed is not readily available in Pakistan.

Local manufacturing, with supply constraints

The project stands out less for any technology described in detail than for how it was carried out. The design and programming were done in the creator’s workshop, while the robots are manufactured in Multan, in the province of Punjab. The report therefore presents a local initiative facing supply constraints, rather than a product with established industrial production or large-scale commercialisation.

Developing the robots locally gives the engineer control over their design and programming. Their continued development also depends on access to parts, followed by the day-to-day maintenance of the machines. The report gives no figures for manufacturing costs or any potential operating savings.

Public interest, meanwhile, already seems evident in the restaurant: staff report that customers come from far away to see the robots and that business has increased. This is an observation reported by employees, with no figures on visitor numbers or sales. Curiosity is therefore a plausible explanation for the immediate appeal, but it does not show that the robots alone improve the restaurant’s financial performance.

Camera vision: a project, not a current function

Aziz hopes to add camera-based image processing functions in the future. He mentions the possibility of the robots recognising customers and remembering their previous orders so that they can make suggestions. These are stated aims, not capabilities attributed to the machines currently in use.

Such a development would also change the nature of the service. Carrying a tray is a physical task; recognising a person and remembering their choices involves processing information linked to customers. Before judging the usefulness of these functions, it would be necessary to know, among other things, what data would be recorded, how it would be protected and whether customers could refuse its collection. The report does not specify these arrangements, so Robia and Jojo should not be described as robots that already recognise their users.

An example of a broader trend, with no evidence of replacement

The International Federation of Robotics (IFR) places service robots across a range of sectors, including hospitality, healthcare, agriculture and cleaning. In a position paper, it presents these machines as tools that can take on certain repetitive, physically demanding or time-consuming tasks, allowing employees to focus on other activities. It also emphasises safety, privacy and reliability as conditions for trust.

This general framework helps put Karachi’s robots in context, but does not validate their specific performance: the IFR provides no assessment of Robia or Jojo here. At the restaurant, their reported function is to move dishes; people are still needed to provide the service, including to take plates off the trays. Nothing in the available information establishes that these machines replace jobs.

For customers, the difference between a demonstration and everyday use is therefore essential. Robia and Jojo make one specific possibility of service robotics visible: moving dishes around a dining space. Their practical value beyond the novelty would depend on questions that remain open about the consistency of their operation, costs and maintenance. For now, the Karachi case mainly shows how a limited task can be automated in an establishment, while the rest of the experience remains shared with people.