We design, build, and manage the most reliable and scalable Cloud solutions for AI Applications
For innovation led enterprises delivering SaaS based solutions, Speed equals competitive advantage. They need on-demand access to scalable development and testing environments to facilitate innovation, accelerate product delivery, and streamline hand-off to production. Our CloudOps Studio speeds up collaboration by combining the most prevalent Cloud technologies (such as AWS, MS Azure) with Continuous Integration (CI) and Continuous Delivery (CD) techniques and together we deliver products faster to market.
Cognitive system for Product authentication in CPG industry
For a US based disruptive start-up, we developed a real time cognitive system to authenticate a range of food products like edible oils and alcohol.
Predictive models to determine authenticity of products were built and deployed in the cloud using ML and AI techniques. The end-to-end system, consisting of a proprietary Raman Spectrometer, a mobile application to communicate with the spectrometer, and a processing system deployed in the AWS cloud, predicts authenticity in near real time. The system collects structured and unstructured spectrometer data through a mobile application to build Predictive Models which were deployed on AWS. The system included a cloud-based Decision Engine that utilized the AWS deployed predictive models to make predictions and communicate results back to the user on the mobile device.
The new system significantly eliminated human-centric workflow to less than 5% of existing levels, and delivered breakthrough competitive advantage over alternate solutions currently available.
Built a Cross platform mobile application including architecture development, core design, UI, and back end processing using Native Android scripts, Angular 2, Java, Python with a Flask micro-service wrapper, for a US based client.
The product was built atop an AWS cloud platform and integrated a front-end Android mobile application with backend processing in AWS. The product was architected to handle large volumes of IOT data as input, run sophisticated Python based predictive models operating under AWS, to derive insights and predictions using proprietary AI and ML algorithms. MySQL was utilized for storing master data in the cloud and MongoDB was the NoSQL database for high volume transaction data for easy scalability.
DevOps for AI
Architected, developed, and deployed a complete software development infrastructure (IaaS) for an AI and ML based system, using tools such as Jenkins, GitHub to automate the development and deployment of software models on test and production environments.
The complete environment was set up on AWS to support stringent availability, processing time, latency, and accuracy requirements, critical to the product value proposition. Additionally, we established a dedicated Advanced Data Science Team and Processes to iterate, scale, and rapidly deploy Predictive ML models in the AWS cloud.