“By 2025, we aim to expand our services across the global market of the US, UK, Europe, South East Asia besides the Indian sub-continent”Sekar Udayamurthy
In 2018, Sekar Udayamurthy, Krishna Iyengar and Vinodh Venkatesan founded Jidoka Technologies to explore the opportunity in automated QC (quality control) solutions.
“Jidoka stands for a principle that advocates ‘intelligent automation’ or ‘automation with a human touch. Currently operating in India, Jidoka serves industry verticals like automobile, pharma, general manufacturing, electronics and printing,” said Sekar Udayamurthy, Co-founder & CEO of Jidoka Technologies.
The Chennai-based start-up offers industrial automation, automated cognitive inspection and cutting-edge engineering solutions. “We positively impact our customers by increasing their throughput by 33%, increasing accuracy by 5-8% and decreasing their false positives by 30%. Our customers get their investment back in less than 15 months,” Sekar added.
Jidoka’s state-of-the-art solution combines the power of AI with automation and computer vision in defect detection, delivering 98% or higher accuracy in the QC process. The QC solution has both hardware and software components. Hardware includes:
- Huron for large volume complex products
- Tigris for flat and lightweight products
- Miyake for manual load and unload with selective checking
Image Credits: Jidoka
Specific models can be tailored to fit the assembly line requirements of different firms. Additionally, the company’s patented software platform Kompass uses AI to create an end-to-end system for visual defect detection in real time. The software provides a turnkey solution that fits seamlessly into the production line to deliver up to 200-220 decisions per minute.
“Our differentiation is that we are an E2E solution provider with proven implementation experience and not just a software platform,” said Udayamurthy.
The company’s core tech-stack includes:
- Ubuntu-based platform
- Docker as the main containerisation solution with C++/ZeroMQ/MERN stack for frontend and backend products
- Core data science stack based on Python/Tensorflow/Nvidia CUDA technologies
- In-house custom tools for MLOps
- Developer tools such as VSCode and Jupyter Notebooks
“Instead of experience, we look for malleability in people who can learn and adapt quickly to the changing dynamics. We have been lucky in hiring the young talent mixed with a few very senior people with experience in architecture and technology to build our products. We also work with Training Institutes to train the talent we hire. Further, we mentor our talent with online and vendor-specific training to accelerate their productivity,” Sekar said.
“We recently received seed funding of US$3,40,000 from a leading technology entrepreneur, and Jidoka is currently valued at US$3 million. We were self-funded while building the core of our product. By 2025, we aim to expand our services across the global market of the US, UK, Europe, South East Asia besides serving in the Indian sub-continent,” said Sekar.
From a software perspective, the company is attracting talent from the US, East Europe and parts of India. On the hardware front, the design, and validation happens remotely, but prototyping and build-outs happen offline. “In the long run, we are looking at options to outsource the build-outs as well, and as a company, we will go through the transition,” said Sekar.
Jidoka is looking at two areas to further product development:
1) Move to a KPI driven triaging of Quality. That is, manage economic, quality and production constraints and drive incremental change rather than looking at quality as an isolated module.
2) The company is looking at the need for customers to have data to ensure higher levels of accuracy. “To do this, we have to leverage the cutting-edge work done in AI/Machine vision such as Anomaly detection, 3D image generation, etc., and to this effect we are productionising some of the concepts while patenting them as well.