Join the Open Network Revolution

Summary.

Technology that has been built over the last couple of decades has been focused on enhancing the lives of a select few with disposable income. Siva 3.0 is what we consider the next stage in the evolution of commerce, built on the best principles of open source, decentralized, collaborative technologies with an intent to democratize commerce and unlock global prosperity and opportunity by empowering every business. We believe that a global commerce protocol is imminent and ONDC heralds more such global projects. Like its namesake who represents creation and destruction at the same time, we envision Siva to represent business both for Good and Profitability. Read our whitepaper to understand how our tech works and how you can build your technology on our tech stack scalably and go to market in no time or why you don’t need to hire a big tech team or depend upon data scientists to run your business and scale it.

The Evolution of Digital Commerce.

Trade has been the cornerstone of human civilization. Historically, the greatest explorers and sea farers have been traders and merchants ferrying their wares and cultures to new lands. For eons, commerce has been tightly coupled – a one-to-one binding of buyer and seller. The nature of physical trade itself restricts the decoupling of the components. However, with the invention of the internet, the paradigms have started to shift. 

The State of India’s Industry.  

Rapid digitisation, an upwardly mobile population and widespread adoption of technical education has propelled India to become the world’s fastest-growing major economy in 2022, by Gross Domestic Product (GDP), at 8.2%, according to the IMF. The Indian economy is further projected to be fastest growing, again in FY 2022-23 with a 13.5% growth in its GDP in the first quarter reported by State Bank of India (SBI). Currently the 5th largest economy with a GDP of $ 3.53 Trillion USD, India is on a trajectory to be one of the top three economic powers by 2035.  

One of the most widely recorded outcomes of the pandemic has been rapid digitisation in almost every industry and geography in the world. A rising tide raises all boats and the Indian ecommerce industry has seen a huge upswing during the period, with experts estimating overall gains to be of the order of 3-7 years of additional growth achieved in that 2 year period. 

• India currently ranks second in the world for the number of internet users, only behind China. Current internet penetration in India stands at 61% or 932 Million internet users, approximately 55% of connections were in urban areas, of which 97% of connections were wireless.
• Third to China and the US, India has an online shopper base of 150 million users in FY2021, and is expected to be 350 million by FY2026.

India’s Digital Transformation

India’s path to digital transformation was first carved, then energized, by important Government initiatives that led to new digital services for citizens and uniformly improved access to such platforms. Aadhar, India’s equivalent of a Social Security Number, was in reality the digital connection to every Indian citizen, covering 99% of the Indian population and allowing citizens access to important government digital services. Not only did this improve the overall availability and transparency of social payments, but it also helped provide timely and effective financial assistance to those in need.

An offshoot of this, UPI has accelerated India’s shift to e-payments, particularly for money transfers between individual parties. According to recent reports, India had 9.36 billion transactions totaling $ 125 Billion USD in the first quarter of 2022. As every transaction is documented and recorded in databases, the likelihood of cash loss and other fraudulent activity is decreased, making it a very reliable option to transact for the various stakeholders within the ecommerce sector, all the while increasing the track and traceability vitally important in online commerce. The National Logistics Policy 2022 aims to to develop a technologically enabled, integrated, cost-efficient, and trusted logistics ecosystem in India to propel industry and commerce. The country’s costs in transport, storage and material handling are estimated to be around 14% of GDP, higher than advanced economies’ average of 8-10%, according to data from the World Bank. The program aims to reduce the cost of logistics to global benchmarks by 2030. 

The State of India’s Ecommerce

The Monoliths.

The Indian ecommerce segment is largely dominated by large marketplaces that operate with a tightly coupled system. These Marketplaces have control over the buyer side and seller side of the transaction, with many also maintaining the supply chain of the platforms as well. During the festive months of 2021, Amazon and Flipkart together accounted for 89% of all ecommerce GMV

Ecommerce platforms like Amazon and Flipkart have been a boon for the industry during the dawn of ecommerce in India. At the time, merchants did not have to start cataloging from scratch and the platforms provided on-ramps to manage catalogs easily. Categorization of products also made it very easy for sellers and buyers to discover each other. On the seller’s end, these monoliths also provided delivery and fulfillment services that were crucial to making ecommerce successful. Merchants also had protection from fraudulent activities. As a whole for the economy, these monoliths reduced the flow of counterfeit goods in the system and also provided opportunities for digital payments to become the norm. 

Customers also heavily benefited from the monoliths as they helped build trust in online commerce. Monoliths also provided easy returns and refund policies, thereby motivating the merchants to maintain quality across their catalogs. The monolithic system also provided reliability of service, fair pricing and cultivating buyer trust, overall.

Logistics and Supply Chain.

Logistics is one of the five pillars that enable ecommerce to function. In India, the logistics and supply chain segment has seen incredible growth, especially post the pandemic. Ecommerce Retail Logistics today is worth $2.2 Billion USD and is expected to grow to $6.3 Billion by 2026. Many Monolithic Ecommerce companies maintain their own logistic and supply chain solutions, Ecommerce shipping has become a lucrative business opportunity. 

In recent years, the emergence of aggregator services have allowed shipping to decouple from the platforms, providing options to the sellers. Logistics players have come up with targeted solutions for first-mile, mid-mile, and last-mile delivery.

The current trend has inspired ecommerce monoliths to decouple their in-house delivery systems and join the open market. Ekart by Flipkart now accepts delivery orders from non Flipkart shipments as well. Meanwhile, Amazon has introduced ‘Buy with Prime’ in the US to allow ecommerce sellers selling on other platforms to leverage the Seattle giants expertise in supply chain management. Meesho recently confirmed the launch of their logistics as a service model while Paytm has been a quasi logistics company for a few years.

The Retail Conundrum.

However, the growth of ecommerce has had little impact on India’s offline retail industry. Limitations are evident from the share of digital commerce in the overall retail segment in India. The Gross Merchandising Value (GMV) for the digital commerce retail market in India was ₹2.85 Lakh Crores (US$ 38 billion) in 2020, which is only 4.3% of the total retail GMV in India and well below the e-retail penetration in countries like China (25%), South Korea (26%), and UK (23%).

India is the 4th largest retail market in the world. The offline Retail industry according to India Brand Equity Foundation is expected to grow at 9% year-on-year till 2030. Some categories in the retail sector – FMCG, Fashion and Beauty – are projected to grow at a rate of 20% and higher. Major retailers dominate the offline retail market. 
Despite their large influence in the retail space, these large retailers account for less than 12% of total retail sales in India. The remaining 88% are considered unorganized businesses. Around 1.2 crore Kiranas (hyperlocal neighborhood provision stores) account for 80% of the retail sector in India, with a majority of them being unorganized, or self-organized and most of them digitally excluded, largely ignored by both large ecommerce platforms and retailers.

The Exposé. 

As of September 2020, India is estimated to have 4.25 crore Micro, Small and Medium Enterprises (MSMEs) that have the potential to flourish with innovative sales and marketing efforts but are not part of this digital revolution. Even on the consumer side, only a small portion (~20%) of the internet users in India are online shoppers. Despite the exceptional growth in India’s ecommerce sector, the pandemic exposed the scattered impact and inconsistent reach of ecommerce in India. Most parts of the retail value chain and supply chain were completely disconnected from the ongoing digital revolution and huge inequalities and disparities in opportunities were revealed. This was especially true in rural areas which covers about 70% of India. Sellers who had inventory, had no access to supply chain networks. Logistics carriers had limited access to inventory, and those who had access to both, could not reach their customers. While the pandemic created massive opportunities to the established ecommerce majors and shoppers in top tiered cities, it also revealed the true limitations of closed-networks ecosystems. The isolation of complete sections of society and retail sectors planted the seeds for shoppers and merchants to look beyond the monoliths, towards omnichannel, de-coupled, equitable solutions

State of the Ecommerce Shopper.

The pandemic had a lasting impact on the Indian ecommerce shopper. Seasoned shoppers became more channel-agnostic. They began searching on Google and buying from various different websites. Those, once isolated, embraced digitization through local e-retail and regional marketplaces.

State of the Ecommerce Seller.

In most countries local ecommerce platforms are competing successfully with Amazon. This is because the demand and supply forces operate in harmony. Merchants, like shoppers, are breaking away from Amazon, and increasing their footprints across varied platforms. However, the pandemic in India left millions of sellers exposed to a couple of platforms. 

Further, the average commission + marketing + FBA fees paid by a successful merchant selling on Amazon is close to 35% of their revenues. This is driving merchants to move to platforms like Shopify which offer a lot more flexibility at a lot less cost. Having said that, merchants on Shopify often face challenges of scale traffic generation, costs of which can account for over 60% of GMV. The solution is Omnichannel – a little bit of everything. The benefits of scale from marketplace platforms, and the control and customer connect from owned channels seem to be the way forward. Merchants selling on 3+ Channels increases order rates by 5x and retailers who don’t sell on multiple channels miss out on 30% of sales

A Brave New #OpenWorld.

The concept of omnichannel has been around for about a decade. Even in the pre-digital era, companies had to consider various channels to sell. The traditional channels would be retail outlets (serviced via distributors), modern trade channels (serviced by companies themselves), B2B channels such as HORECA, and other special channels (e.g., Armed Forces). However – these channels were independent, and rarely were synergies applied across channels. 

In the digital era, the role of omnichannel is to integrate the various channels (offline and online) to provide a unified buying experience to the end customer. Until recently, one of the critical roadblocks to omnichannel adoptions was the self-sufficiency in the monolithic structures of Amazon and Shopify. The pandemic changed all this. For Shoppers and Sellers – one channel was just not enough.

Taking cognizance, Amazon and Shopify are both taking strides towards omnichannel enablement; the once tightly-coupled monoliths are opening up their technologies and platforms to customers and competitors alike in a bid to transition to open, interlopable and composable technologies. A further evolution of open networks and composable commerce, platforms are also embarking on technology commercialisation and creating additional revenue streams by lending access to their tech stacks for smaller merchants to leverage.

Let’s take a closer look :

Amazon’s omnichannel journey is centered around its grocery business. Amazon has realized that to sell groceries successfully; they need physical stores. Thus, the launch of Amazon Fresh – A physical retail chain powered by Amazon. A quick reading of Amazon’s history will tell us that groceries are merely a pilot. Eventually, they will likely sell most categories through the multichannel route. 

What we find most interesting is that Amazon recently launched a major initiative called Buy With Prime. With this new initiative Amazon is telling merchants that irrespective of which platform they are selling on (Shopify, Bigcommerce, etc) or what kind of sales (retail, ecommerce, B2B, etc), they can simply sign up for Buy With Prime and Amazon will drive demand for them through their marketing channels and fulfill their orders through FBA.

Amazon’s Multi-Channel Fulfillment (MCF) program is one of the fulfillment options Amazon offers to sellers. By storing your inventory at Amazon’s warehouses, they’ll handle fulfillment for your brand across select channels – including shopify and other competing marketplaces. Unlike Fulfillment by Amazon (FBA), where only Prime orders from Amazon’s marketplace are fulfilled, MCF uses a combination of shipping speeds and not just the famous 2-day shipping.

Prime customers can now search for products, find various options near them, such as a local store or a webstore that is not Amazon or some seller on Amazon’s marketplace, place their orders with the same convenience that they get by being Prime customers on the Amazon marketplace.

This has the potential to make Amazon indispensable to brands as they become the bedrock of commerce irrespective of where the sale happens.

Shopify Plus’ omnichannel module allows brands to sell to 80+ countries. To enable the same, Shopify offers 70+ Shopify themes to customize their frontend look and feel and generate great user experience across devices and is partnered with 100+ payment providers. Furthermore, Shopify lets users create custom multichannel automation flows using Shopify Flow. Brands can now automate their customer journeys easily without causing breaks in the user experience. 

Further, Shopify’s Hydrogen and Oxygen platforms (recently launched/being launched) allows companies to compose complex integrations and systems on Shopify, literally giving wings to what is now possible to build on top of Shopify’s already extensive suite of solutions. Shopify’s playbook is what every other platform on earth aims to build. What makes them special is that they already have built it.  

It appears the tides are turning and the shifting balance of power, in favor of the shopper and merchant, has triggered certain advances at Shopify and Amazon – with both spearheading towards openness, technology sharing and commercialisation. This creates some exciting possibilities for mid-market brands and merchants. Using Hydrogen and Oxygen, brands will be able to compose future-ready systems. We can see cases where the brands leverage the openness of Shopify to expand to multiple geographies and fulfill orders using Shopify’s new fulfillment capabilities or MCF/FBA. Similarly, we could see instances where products are advertised online, ordered on Shopify stores, but sold via Amazon’s physical stores and vice versa. Platforms are gradually becoming more interoperable.

The Dilemma for Brands and Merchants.

The cost of loyalty. While there are numerous advantages for brands in using platform innovations such as Amazon’s BWP or MCF program, it comes at critical costs. The Wall Street Journal recently reported that Amazon uses sellers’ success and customer data to spin off its own private label products. The fear grows. 

Brand’s have no access to their own customer data, no system for remarketing or creating loyalty, no avenue for word of mouth marketing or referrals. Between the fear of platforms creating competing products, and the apparent disconnect with their shoppers, brands are wondering if they should stick with Amazon, switch to a 3PL, or manage multiple sales’ channels and fulfillment in-house.

Today, sellers and buyers around the globe need a system that genuinely makes ecommerce plug and play for small businesses while offering larger businesses the ability to build vertical solutions for their custom use cases, allowing developers to build both for themselves and for the community. However, no existing system offers businesses the best of all these worlds. 

The solutions to run ecommerce businesses successfully exist, but in a fragmented manner. A comprehensive unified solution cannot be offered by one player. Thus, we see the industry pivoting to an open-sourced model. Open source brings together communities focussed on collaboration and interconnectedness and commerce today is the latest industry moving to the open source model.

What is Open Source?

Open Source is a global movement amongst developers that promotes the development of decentralized software through open collaborations. While many non-developers misconstrue open source as free to use or commercially free technology, Open Source is much deeper than that. 

Open source is a philosophy that stems from the idea of community over code. Originally the movement began as a response to proprietary code, or software or technology or knowledge owned by corporations and not accessible to all. The movement today can be defined by the different participants in different ways but generally, all agree on 3 principles: 

• To not allow profit motives to restrict access to life-giving technology
• Making patents accessible for those who are engaged in a humanitarian activity for free
• To not allow patent law to slow the pace of innovation or service delivery to the needy under any circumstances

The emergence of a global commerce protocol – ONDC

The government of India initiated a pilot project to level the playing field for businesses of all sizes. DPIIT (Department for Promotion of Industry and Internal Trade) and MCI (Ministry of Commerce and Industries) together envisioned an open network to unbundle and facilitate digital commerce at a population level. Inspired by previous government initiatives, like UPI and UIDAI, the Open Network for Digital Commerce (ONDC) is an initiative that aims to democratize and integrate Digital Commerce, migrating commerce from a platform-centric model to an open network. ONDC will eventually connect all transactions digitally, enabling local buyers and sellers to trust and transact through an open network, irrespective of the platform/ application they use. The aim is to help small local buyers and sellers to create a niche identity on the network and conduct their business, without enrolling on several sites. 

Unlike a Platform that is a closed loop with high barriers to entry, a protocol is a set of rules made to govern a network. The ONDC Network Protocols simply allows buyer side applications like ecommerce stores, travel sites, food delivery applications, edutech, etc to work agnostically with the seller side participants like kiranas, B2B vendors, logistic players etc. to complete the transaction. The protocols maintain registries to allow participants on to the platform and also ensure the network policies are maintained by all parties. The ONDC Protocols are also built to tackle the 4 steps of the transaction – the discoverability of the products and services on the network, the transaction between parties, the fulfillment of the contract and the post sales arbitration.

Zooming out to see the holistic view, a fully implemented ONDC will bring a paradigm shift across the Indian ecosystem. A Kirana Store operating on the ONDC will be able to sell directly to residents in their vicinity, while buyers will finally be able to buy across all platforms at the same time. The ultimate implementation of omnichannel in theory. The system will finally enable customers accustomed to ecommerce to discover micro and small scale unorganized businesses like small carpenters, cobblers, electricians and repairmen. The network will trump the discoverability barrier beyond buyer apps and provide cross platform visibility. Sellers will not be locked into the ecosystems of big platforms and be able to port their reputation beyond any single platform.

The need for an application layer

ONDC will address the most common issues faced by sellers and buyers once implemented. However, the implementation of ONDC in a complex ecosystem such as commerce is unlikely to be easy. ONDC is a first-of-a-kind initiative being implemented at a population scale. The implementation of ONDC will bring millions of small merchants hitherto ignored by the progress of technology into the organized sector and enable them to be discovered and trusted by the population. Clearly, ONDC will bring about new challenges of its own into play.

Solution for End users

The biggest chunk of the sellers who are not present online is because there is a huge barrier to entry for small merchants like a standard mom-and-pop store. The listing of products online on any eCommerce website is a very comprehensive and tedious process. One needs to upload the product images, go through various onboarding forms, upload documents, provide the correct title, keyword etc. For a small seller (eg a costermonger) who might not even know that goods can be sold online the biggest hurdle is the onboarding on the platform. 

Current platforms have a large number of mandatory steps that need to be completed which small sellers might not be able to complete independently and with ease. Sometimes these sellers are also trapped into paying for these onboarding services by middlemen. Siva onboarding simplifies the entire process of onboarding for the seller and makes it easier even for a costermonger to start selling online. Siva accomplishes the entire onboarding using Data Science algorithms, cataloguing and price prediction. The entire onboarding journey is done via Whatsapp thus making it platform independent.

Macro Economic Impacts

The question of mass-scale deployment of the open network platform app comes up. How do you convince millions of people to install a new application on their phones? Many buyers/sellers from rural parts might not even know how to install an application. During the covid pandemic, it took months for the government to make people download and install the CoWin application on their phones. 

We try to use the current ecosystem that has enough penetration. So we developed a chatbot that can sit on top of any messenger application and help the end user interact with the platform. There are about 49 crore Whatsapp users in India. Here we have solved the economic case of finding the cheapest way to have the open network application be mass deployed. This saves the government crores of rupees in rolling out the network and provides for better adherence and lesser cost of training. Can we help users (like the local mom-and-pop store or the costermonger) upload prices, and catalogues of their products via WhatsApp messages..and get and place orders via WhatsApp?

There are about 40 crore Instagram users. The majority of them create reels which showcase a product or service. How can we leverage those reels to create product catalogs? People are using Instagram to sell their products as part of social commerce. How can we onboard them to an open network?

If we make onboarding easy, if we make network participants work easy via our iPaaS, if we make our development community work easy and if we utilize the currently available extremely wide-scale ecosystem we can create a massive impact in the online commerce space not just in India but around the world.

Can we one day go and take picture of a product or service and then upload it to the open network or convert a video of someone buying a shirt and push it onto the open network? What will be the overall network economics? Who will do what activity? Who can do what activity? These are some of the questions that we need to figure out the answers to.

Onboarding can be broadly classified into seller onboarding and buyer onboarding. The buyer-side onboarding is pretty straightforward. The three touchpoints for the buyer could be existing buyer-side apps like Flipkart, Meesho, Amazon etc or it could be directly via chat/voice through messaging apps like Whatsapp or it could be via Siva buyer-side apps. Post authentication and digital ID creation the buyer can place orders via the order management module (Fermion). The ID, authentication data and orders are then passed on to the open network via the Siva connector (Boson). 

The seller side onboarding also has three touchpoints, to begin with. They can get on board via existing seller-side apps and POS systems like Snapbizz or messenger services (SIVA Data labs chatbot) or Siva buyer-side apps. Once the authentication and digital ID creation are done the seller can start creating a product catalogue via text, voice or image or directly on the Siva seller side app. The cataloguing module of Fermion and Boson will help to create SEO friendly catalogue and will do most of the heavy lifting for the seller. The Many Worlds forecasting module will also assist in determining prices for the uploaded products depending on whether they fall into the branded or unbranded categories. Once the products are catalogued they get broadcasted into the open network. From this point on the order gets routed within the open network. The seller can opt for own shipping or can use third-party shipping in which the seller becomes the buyer (for the logistics chain). The payment for the actual order can happen using the seller’s independent payment network or via the Siva connector (Boson).

The biggest challenge is to onboard at scale. For that, we currently do not have omnichannel-enabled product taxonomy which can be treated as a uniform one. Also, we have to ensure that we have pricing for that locality ready to ensure that the onboarding is hassle-free and at scale.  The wide variety of product taxonomy and price differentiation for every small pocket of our geography.

Solution for Network Players

A network is only as strong as its weakest link. Therefore, to bring forth a strong and powerful network, we need as many players as possible. The real innovation in technology comes when multiple players join together to create a unified system that helps to solve common problems.

Two separate systems that work towards solving the problems of a common subset of people can work wonderfully well. But if they are combined and work as one entity the end solution and its benefits are multiplied.  We have an iPaaS solution that makes it very easy to integrate any existing eCommerce solution as long as they have APIs for integration.

We are loosely coupled system and domain agnostic. Thus we don’t have a walled system which only allows limited participation. Our targeted use case is retail, eCommerce and supply chain but our system is built to accommodate all players. If you think you can solve even a minuscule problem either for the buyer or the seller then you are a perfect candidate for becoming our network player.

We have ready integrations with some of the biggest companies in the domain. This allows others to use us as a springboard to create a larger network ecosystem to solve the most complex problems faced in eCommerce today.

Domains we have picked

Marketplaces: Brands can expand their global footprint by effortlessly syncing their products, managing their orders and increasing their sales velocity.

Webstores: Brands can sell exclusively on their own custom website without the competitor’s products vying for the attention of the buyer.

POS: It unifies one’s online and retail stores to strengthen sales conversions everywhere one sells, streamline store management, and help one build stronger relationships with one’s customers. It has all the essential features one needs to make sales, track performance, and manage customers, orders, and inventory.

Logistics – These provide leading end-to-end tech-enabled logistics solutions with cutting-edge technology and automation to enable first-mile pickup, processing, network optimization and last mile delivery.

Third Party Fulfilment – 3PL allows brands to leverage the power of being where their customers are. 

Payment Systems –  With payment systems integration, the brand’s customers can now have a Digital Wallet for all the transactions like purchases, cashback etc right at their fingertips. It provides for paying for an order easily and gives a seamless checkout experience for the buyer on the platform.

Accounting Softwares – Sellers can save time and stay ahead of their bookkeeping with an accounting integration. They can track income and expenses and maximize tax deductions.

Marketing & Analytics – Marketing and analytics help sellers provide personalized recommendations to their buyers, lets them optimize prices and allows them to know their customer better.Project Management Tools – These help sellers increase efficiency, boost performance, resolve customer issues and ultimately improve customer satisfaction.

Build your own integrations

An iPaaS, or integration Platform as a Service is a platform that standardizes how applications are integrated into an organization, making it easier to automate business processes and share data across applications. It not only allows for easy integration and communication between an application and the platform but also allows for easy information sharing between different applications.

Siva comes with a low code and no code iPaaS connector capable of connecting any application with open APIs to the platform or any other Open APIs application via the platform. It gives the user the ability to choose the JSON they want to connect and filter at the metadata level. Now any custom solution will be just a few clicks away.

Open for Developers

As demonstrated earlier, a Network Protocol Layer is a set of rules and policies that govern a system. These Protocol Layers have frameworks, rules of governance and resolutions, and policies that are vast and complex. Like any Protocol Layer, ONDC is also quite complicated. For users to interact and work with any protocol layer itself would be quite complicated and difficult.

Siva Fermion – Core

Core Architecture

In the Data Flow Diagrams let’s start with the developer (bottom left). He/she submits an app to the listing app store. If the app is approved the secret key gets generated and if the app is rejected it gets back to the submit stage. Once the secret key is generated we put the app in the Siva app core repository. We keep a few details of the app like 

• App Name
• App Technical Name
• App version
• Description
• Developer details
• Added component list
• Secret key
• Code Base
• Concurrency list
• Access class
• Security group
• Cron job
• Database schema modification
• Dependent apps
• Third part connector list
• Micro front end
• Micro back end
• Purchase History
• File system

On the user side they first access the Siva frontend. There can also be a third-party app frontend which comes from the app store (where the purchase of the app might happen). There are also different microservices which render to the Siva frontend. There is a core component library that gets to the Siva frontend core and gets delivered to the user. From the front end, we check if the data is transactional and then we move it to EDA if not we move the data to the Redis cache. Even from EDA, we can push to Redis cache. 

Then there is the Siva core backend engine. From the AppStore, we create lots of libraries like the concurrency library and the connector library, the cron library, the access class library, the DB schema library and the codec file system library. From the concurrency library, we have a Siva concurrency core which handles data with the backend engine. In the same way, we have a connector library that gets connected to our Siva connector and from there goes to the mapper service and then to connector EDA and finally to third-party applications. Its a bidirectional data flow. 

Similarly, we have the cron library which connects to the cron core and that connects to our database. Then there is an access class to the access class core which finally connects to the core engine. In the same way, we have DB schema to ORM core code and from there it goes to the SQL, NoSQL database and also gets connected to the core engine. 

In the Siva PII manager, we manage all the PII details. We adhere to the PII specifications of different buyer-side and seller-side apps. If someone requests to delete the PII data after some time the cron ensures the data is deleted. It also ensures that the PII lives up to the overall network expectations since the buyer side can have their own PII requirements and the seller side can have their own PII requirements. The cron manages the PII data which makes it compatible with the compliance norms. 

From the app store, we have the core and file system library which connects to the siva core and file system core functionality which finally gets connected to the file system The siva code and file system library is then connected to the Siva GO compiler via a check (if not it goes via a GO python compiler). This finally goes and sits in the backend engine.

Core Backend Architecture

The back-end is the code that runs on the server, that receives requests from the clients, and contains the logic to send the appropriate data back to the client. The back-end also includes the database, which will persistently store all of the data for the application

Developers can use the Eunimart Platform on their own servers. They can acquire the Backend Core code from the Siva3.io website and customize the code as per their requirements and host it on their own server. They can then connect it to their other solutions using Open APIs

Core Frontend Architecture

Micro Frontend Architecture

Micro-frontend architecture is a design approach in which a front-end app is decomposed into individual, semi-independent “microapps” working loosely together. The micro-frontend concept is vaguely inspired by, and named after, microservices.

Today, the Monolith Frontend is being phased out in favour of Micro Frontend Architecture, which provides:

• Scalability to allow rendering of multiple modular frontends
• Heterogenous frameworks for running in a single application
• Avoids complexities of a large monolith
• Allowing multiple teams to peacefully coalesce their development efforts
• Separate release schedules/delivery timelines

Highlights

• Eunimart’s PaaS system have different microservices running on different ports to achieve scalability & reliability. 
• To add safety, Eunimart is using Nginx Reverse Proxy [web server] to redirect frontend URLs and for configuring the SSL.
• We are using Node.JS to host frontend applications and to make them run on defined ports. 
• We are using PM2 to host this node server which gives the advantage of a load balancer for different cores of our server. 
• PM2 enables you to keep applications alive forever, reloads them without downtime, and helps you to manage application logging, monitoring, and clustering.

Siva Boson – iPaaS

Siva Mapper – iPaaS

An iPaaS, or integration Platform as a Service is a platform that standardizes how applications are integrated into an organization, making it easier to automate business processes and share data across applications.

The customer can integrate their 3rd Party Solutions with the Eunimart Platform using Open APIs and connecting them using our iPaaS.

Siva Many Worlds – AI and Data Science

Cataloging Architecture

Onboarding comprises multiple subsystems working together seamlessly. First is the chatbot which allows communication and transfer of data from the end user (seller in onboarding scenario) to the Siva platform. Then comes the cataloging part which comprises category mapping, template mapping, field prediction, image optimization, keyword generation and description generation.

The sellers can connect to third party seller apps, the messenger apps like Whatsapp, and Facebook messenger or they can use the Siva seller app. The buyer can either use the messenger apps or use the Siva buyer apps. All the requests from the buyer and seller come to the Siva EDA (Event Driven Architecture). After passing through EDA we first check if the product query is clean (ie has all the data points). If the query is clean it goes to the category predictor and then to the Siva catalog repo process. That connects to the price predictor and the demand forecast processes. 

If the buyer/seller side app schema is available then it goes for a query in product taxonomy. If not then we do a Siva AI schema mapping. Finally, it goes to the EDA and the fields get predicted. If not we manually do a schema development and add it.  

If the query is not clean it continues to CNN operation which queries only with image, voice or text. That gives the output as a clean query which again passes through the check. The query recirculates within the CNN till it’s clean. In CNN if we have the product image available then we check if the voice is available and if it is then we convert the voice to text.  We provide this text (without the voice) to CNN. If originally the image was not available then we check if the text is available or not. If the voice is not available then again we check if the text is available or not. If the text is not available then again we go for a voice query.

The elastic search goes for a product query inside the Redis cache where the product taxonomy is stored. Then we find if we have the image available for the whole thing or not. If the image is not available then we put it into EDA and ask for the image from the user. If it is available it goes to CNN. 

The catalogue predictor works as a part of the catalogue repository engine. The repository also works with a virtual try-on engine (VITON). There is a deep neural network in the predictor which makes the predictor work better the next time. The Siva catalog repo is connected to the Siva EDA (which is the same as the top/original EDA)  There we go for a query to check if it’s a branded product. If yes we put it back to the EDA and send it back to the Siva brand connector API. Here we get the data from brands or brand product databases. If its non branded then we go for Siva non-branded catalog generator. For this, we use Siva NLG (Natural Language Generator). We provide the data back to the catalog repo. In the image storage database, we do a Siva Image and Classifier cron job so that whatever image we have we keep labelling it. This helps when we get a query from catalog repo we can provide with the image search and that gets transferred to the product taxonomy.

To read the full whitepaper, download the Siva Commerce 3.0 whitepaper from the top of the page.