Your FinTech Guru

This is a blog dedicated to exploring the vast subject of Finance and understanding Finance Technology (FinTech). I'm a software engineer and MBA (Finance). As a tech enthusiast I'm fascinated with the idea of integrating IT with finance.Besides I love playing video games, guitar and developing games.

[kunaljethwani]

What is Web3 and its connection with Blockchain? Tech Layoffs Analysis, Recession and more

Tune in to my podcast to find out, now available

On Spotify

Agile

Check this out! Listen now on @Spotify.

On Amazon Music #PodcastsOnAmazonMusic

On Google Podcasts

Agile

Confused or curious about that latest technology? Wish somebody could explain that in short and in simple terms? Want to know interesting pr

[kunaljethwani]

Read more of my articles here

While I have got more active recently in penning down my thoughts about this space, I have housed them at a new location for now. You can read more of my articles at my new space - 

linkedin.com/in/kunaljethwani

Back when I started this personal blog over a decade ago, things were different. IT was just the beginning of social media as we know it and we were still in Web 2.0 proper. Linkedin wasn’t what it is now and my presence there wasn’t the same. Over time, my knowledge and skills have expanded yet I have come to realize the time constraints so I want to consolidate my leadership thoughts at a singular place, for now. 

[kunaljethwani]

It always seems impossible until its done.

After what seemed like a long hiatus from blogging, I’ll now be starting to post exciting reads more frequently. Watch this space for articles on RPA, chatbots, conversational AI, ML, VR and a lot more. These will be be short to medium reads or a series of articles on a certain topic related to these technologies and their implementation in the real world. Follow the blog to never miss out on any post.

[kunaljethwani]

Mortgage Lenders Should Take Hints from Spotify

What do mortgage lending and music streaming have in common? More than one would think.

In fact, mortgage lenders could stand to learn a few lessons from how music services such as Spotify interact with consumers, according to Vishal Garg, founder and CEO of Better Mortgage.

Better, an online mortgage lender utilizing machine learning to streamline the lending process, has already incorporated some of Spotify’s ease of use into its platform (it helps that the company’s chief technology officer is Erik Bernhardsson, former head of machine learning for Spotify).

“If you think about what [Spotify] is, it’s ‘here’s a catalog, tagged with attributes, and consumers with preferences,’” Garg told Bank Innovation. “It’s a large-scale preference to attribute matching [system], and that’s what the mortgage process is effectively as well.”

Better Mortgage, which has 21 of the largest mortgage investors on its platform — accounting for approximately $1.2 trillion of annual mortgage demand, according to Garg — allows users to know instantly what they can afford, by using machine learning tech to match consumers (and investors) with the right products.

Creating a Spotify-like experience for mortgages brings a greater degree of transparency and ease to the consumer, while also allowing the company to lower rates. Better is so confident it can give consumers a less expensive rate the process now comes with a guarantee; either Better beats a competitor offer by at least $1,000, or it will give the consumer $1,000 in cash.

The reason mortgages, one of the most competitive markets in the world, seem so frightening to consumers (especially those who are purchasing a mortgage for the first time) is because consumers are shopping for a house at the same time, according to Garg.

“We’ve made it an online process — you walk into an open house, you pull out your phone, give us a few details, and we can approve you [for a loan] for that house while you are still in the house,” Garg said. “It takes the entire stress out of the home-buying process, because you already know whether you can afford it or not.”

[kunaljethwani]

Small and Challenger Banks ‘Increasingly’ Integrate With TransferWise

When TransferWise first launched, CEO Kristo Käärmann did not expect a favorable response from incumbents, which still charge consumers considerably higher fees for cross-border transfers, compared to TransferWise’s 0.5% rate (in the U.K., on average).

But the opposite trend began happening.

“We learned overtime that banks have an incredibly high cost-base to do the same thing as we do,” Käärmann said at the Future of Fintech conference this week. “It turns out, it costs them seven times more to do a foreign exchange transaction than it costs as.”

So instead, small regional banks, as well as neobanks, began integrating Transferwise into their apps as an option for international transfers.

“There was this small bank in Hungary, which would help its customers onboard on TransferWise,” instead of using their internal correspondence service, he said. “We see that with neobanks like N26 and Monzo; they are not integrating with Deutsche Bank for correspondence banking, they are integrating TransferWise.”

N26, for example, allows users to directly link their account to TransferWise for international payments. “It’s like ‘Login with Facebook’ or Google, and it’s exciting to see this in banking,” he added.

Currently, about $1.5 billion goes through the company’s system monthly. “But the number I am really proud of is that our customers save $1.5 million in transfer fees daily,” Käärmann said. “In the U.K., we charge an average of 0.5%, so $5 to a $1,000. Banks usually charge $25 of international wire transfer fee,” in addition to the average 4% rate. “So, for $1,000 it comes up to $75.”

The company is already surpassing traditional banks in market share, at least in the U.K., according to Käärmann: “They [banks] should do the same thing as neobanks do; that’s the right thing.”

[kunaljethwani]

A SoFi Mortgage Comes with Avocado Toast

Millennials out there are often caught between two big decisions: do they buy a home, or do they keep buying sweet, delicious avocado toast?

Luckily for the potential millennial homebuyer, fintech lender SoFi has found a way to merge these two completely exclusive purchases: a SoFi mortgage now comes with avocado toast.

Seriously—for the entire month of July 2017, anyone who buys a home with a SoFi mortgage will receive a month’s worth of avocado toast, delivered to their home. Homebuyers can even decide between regular or gluten-free bread.

The company posted on its blog yesterday:

Why avocado toast? Because with a SoFi mortgage, you don’t have to skip out on the avocado toast while saving up for a down payment. SoFi mortgages make it possible to buy a home with just 10% down and no borrower-paid private mortgage insurance required, which could get you into your dream home sooner. It could also make it possible to do that while brunching.

Sofi also offers potential home-buyers “debt-to-income” mortgages, which provide more flexibility to applicants that might have a higher student debt.

A home, plus a daily delivery of green goodness on toast for a whole month! Don’t get too excited though — potential home buyers will still have to actually toast the bread themselves for the “full experience,” as SoFi was sure to warn on its blog.

But hey, even millennials can’t have everything.

[kunaljethwani]

Roboadvisors Need to Find the Right Amount of Automation for Consumers

Does your roboadvisor have too little automation, too much, or just enough to satisfy consumers?

That’s the question robodavisory services should be asking themselves, according to Matt Wilcox, senior vice president of marketing strategy and innovation for Fiserv. The reason is that while automated investing services are becoming more mainstream in financial service, it’s the level of automation that decides whether a consumer is actually comfortable in using them.

“With robo-investing, it’s going to be a fine line,” Wilcox told Bank Innovation, especially as roboadvisors learn to give more focus advice in order to properly manage larger portfolios. “When there’s more risk associated with it, we’re going to find the comfort level of people.”

Read more

[kunaljethwani]

The ultimate 3500-word guide in plain English to understand Blockchain

[kunaljethwani]

[kunaljethwani]

RBI may introduce tough riders for P2P lenders

[kunaljethwani]

Life insurance companies are in early stages of discussions to see if access to ones social media profile could be an enabler to offer better rates of premium for a policy.

[kunaljethwani]

Indian IT (information technology) industry might like to put all the blame on Donald Trump for its woes single-digit growth, retrenchment and reduced hiring from campuses. But the Trump effect masks the crisis that is silently brewing.

[kunaljethwani]

5 already prevalent trends in Fintech to look out for in 2017

2017 is on the horizon and it’s bringing some strong focus on these trends along with it. Following are some trends in the Fintech space that everyone is keeping an eye on right now, and they may very well become the fundamental pillars of financial services in the next decade.

5 Biometric Authentication

No matter how much you upgrade your security measures, there is always this other side of the force that finds its way either through or around the security. Hackers and other criminals can bypass most measures with relative ease, exposing sensitive user information to the wrong people. Biometric authentication has the potential to change that. It may only be a matter of time until we use fingerprints and iris scans to access financial services. Voice authentication is still an option as well, although its feasibility has yet to be determined.

4 Blockchain Technology

Most of 2016 saw a lot of pilot projects and Proof-of-Concepts happening around the disruptive technology Blockchain.

The success and viability of blockchain technology in the real world is yet to be determined. We will see blockchain adoption in the real world sooner rather than later as many banks and financial institutions prepare to embrace this technology in 2017.

3 Big Data And Smart Data

Big Data has been a buzz-word for a while now but its gaining more prominence by the day in the Fintech space. Companies have a treasure trove of information at their disposal, yet fail to harness it. That situation is slowly changing as big data analytical tools are becoming more accessible to everyone in the world.

However, big data will not be the ultimate answer to identifying customer needs and desires. Smart data, which defines big data into tidbits that are easy enough to be interpreted and linked to specific use cases, is the new norm. The largest and oldest banks in the world are the most suitable candidates for this technology as they have huge chunks of data lying around unused to its maximum potential as of now.

2 Mobile Services

It is impossible to deny the opportunities that mobile solutions offer both the unbanked and banked people of the world. Since we carry mobile devices at all times, the world is just a few taps and swipes away. Finance is one of the primary points of focus during the mobile revolution. Digital wallets, messaging platforms, and various products are all targeting a mobile crowd. Rest assured that this trend will continue for many, many years to come.

1 Artificial Intelligence - Deep Learning

The biggest advancements are being made in AI and deep learning right now. With artificial intelligence creations being capable of learning new things themselves, significant steps are taken to create some form of consciousness. This is not to be confused with the earlier forms of automation that made a lot of things simpler and easier for us including the chance of losing a job. It is only a matter of time until AI takes over some (more) of our jobs, but, well…what not for the betterment of our society as a whole, right?

[kunaljethwani]

Artificial Intelligence Will Suffer From Bias In The Same Way Humans Do

Artificial intelligence is often touted as the technology that will eventually reduce human error to zero. And why not? After all, this technology is capable of Moreover, this technology will be capable of taking over mundane tasks, allowing the human brain to be used for other important duties.

However, contrary to popular belief, artificial intelligence can be affected by bias– a rather troublesome aspect to think about, and one that definitely should not be overlooked.

Artificial Intelligence Bias Is A Real Thing

Even though the concept of artificial intelligence is in early stages of development there is a lot of concern amongst the experts about the possible pitfalls that this technology brings along to the table. However, that does not mean that AI has no bright future ahead. 

It is important to take these concerns into consideration, though, as they will change the way we think of AI in general. AI relies on the data it is fed and it is right to say that it can is molded by the data it is fed just like correct nutrition helps our body develop in the right way. Now imagine what if the data fed in is biased already? The technology will automatically assume things are correct! Getting off on the wrong foot is never a good start, and even though AI is capable of learning, some hand holding will be involved during the early stages of deployment.

The way we interact and interface with artificial intelligence will help in the learning process. Interaction is a critical part of evolving AI into a better solution, but even interaction can form a sense of bias. The Tay AI project launched by Microsoft is a great example, as its interaction with the real world turned a straightforward “program” into a very racist communication tool.

But perhaps the biggest threat to artificial intelligence is how it will be affected by human bias in general. Developing an intelligent cognitive system is one thing, but making it objective is rather difficult. As people are involved in the development of these new tools, there will undoubtedly be some form of bias involved. 

Developers have particular preferences, mannerisms, and manners of communication. Those traits will reflect on the AI itself, although they may be phased out over the course of several years.

It is evident that artificial intelligence will have a very bright future, but it will need to go through several iterations before it can be looked at as a completely separate entity. That is not necessarily a bad thing, as letting these creations loose upon the world without proper guidance could have catastrophic effects. Bias will be bias, an area in which humans and AI are not all that different.

[kunaljethwani]

Smart Contracts: Benefits and Use Cases

What are Smart Contracts?

Almost all prominent blockchain platforms now extend their capabilities in the form of smart contracts. A general definition of a smart contract  is a computer programme that can automatically execute the terms of a contract. They extend the utility of blockchains from simply keeping a record of financial transaction entries to automatically implementing terms of multiparty agreements.

Self-executing Systems

Blockchain-based smart contracts are self-executing. They can solve the problems of counterparty trust in the sense that they automatically implement the terms of an agreement between parties on pre-set logic without the need for intermediaries. They are executed by a computer network that uses consensus protocols to agree upon the sequence of actions resulting from the contracts’ code.

Coded contracts introduce efficiency of automatically generating contracts based on mutually agreed-upon patterns and syntax amongst counter parties. This is a major overhaul from how things are currently done i.e. manual documentation. Prior to blockchain, for an agreement of this type, parties would have had to maintain separate databases. Blockchain however, allows the shared database to have self-executing smart contracts where all participants can validate the outcome instantly without requiring an intermediary.

  Best-Fit Scenarios

A good fit for blockchain enabled smart contracts could be a scenario where frequent transactions happen between a network of participants and manual mechanical tasks are performed repetitively for each transaction. Smart contracts are particularly well suited for the permissioned/private blockchain network. For the financial and securities sectors, such a code-based compliance would save a lot of time and money. Syndicated loans are a $4 trillion plus market that still run primarily on faxes, emails and Excel spread sheets. It can definitely see improvement with this technology.

  Benefits

Blockchain based smart contracts offer many benefits for a wide range of applications.

Lesser risk

Decentralised process of execution eliminates the risk of manipulation, since execution is managed automatically by the whole network, rather than an individual party.

Real-time and accurate

Smart contracts use software code to automate tasks that would otherwise be accomplished through manual means. Hence, they increase the speed of business processes and are less prone to manual error.

Fewer/Zero intermediaries

Smart contracts reduce or in some cases can completely eliminate reliance on third-party intermediaries that provide ‘trust’ services such as escrow between counterparties.

Lower cost

Less human intervention results in reduced costs.

New business models

Smart contracts, through their lower costs for ensuring reliable transactions, enable new kinds of businesses like automated access to vehicles and storage units. This can be in conjunction with other emerging IT trends like IoT and Deep learning.

  Blockchain Smart Contract Use cases

[kunaljethwani]

Four Key Blockchain Platforms: Their Use Cases in Financial Services

Blockchain, for the most part, no longer needs an introduction. So, I’ll be focusing on some of the most prominent blockchain platforms and how they apply to different scenarios.

Before moving on to the comparison, I’d like to explain some key terminology:

Proof-of-work (POW) system is an economic measure to deter denial of service attacks and other service abuses such as spam on a network by requiring some work from the service requester, usually processing time by a computer.

Proof-of-stake (PoS) is a method by which a cryptocurrency blockchain network aims to achieve a distributed consensus.

Ethereum

As of now, Ethereum uses a POW system for validation. However, it may rely on PoS at some point in the future. This method of validation requires holders of Ether to prove how much they hold, provided they meet the Ether quota set by the Ethereum Foundation. Using a method of what’s known as weak subjectivity consensus, the nodes validate the network and are paid out a percentage of newly created Ether in return for holding the currency.

Potential use cases range from Decentralised Autonomous Organisations (DAO), to Crowd prediction markets such as Augur. Given the flexibility of implementing smart contracts, most use cases suited to smart contracts are feasible.

Ripple

Ripple has positioned itself as a SWIFT 2.0 RTGS system. Its native currency XRP is positioned as a bridge currency among asset pairs on the Ripple protocol, in order to save money for financial services institutions. Transactions are confirmed every few seconds through a consensus among validators i.e. the Ripple Consensus Ledger. Validators are not rewarded for their part in achieving a consensus. Since Ripple does not use POW or PoS, it can achieve a much shorter timeframe for each round of confirmation in the ledger.

Ripple is naturally suited for cross-border payments. Other use cases make use of Ripple wallet balances to pay in bitcoin and trading bullion against fiat for profit.

Hyperledger

Current implementation uses the Bitcoin UTXO transaction model. It ensures that unspent transactions remain uncompromised while travelling to their destination, through a system of private and public keys. This system ensures that Hyperledger does not have any instances of double spending by controlling outputs. No native currency exists. Also Chaincode (Smart Contracts) functionality increases the flexibility of the platform in terms of implementation.

Syndicated loans and capital markets infrastructure are among the top use cases for this technology.

R3 Concord

R3 implementation is somewhat different as it achieves consensus at the level of individual deals between firms, not at the level of the system. Transactions are validated by the parties to transaction, rather than a pool of unrelated validators. There is also no native cryptocurrency.

R3 is implemented in two parts:

‘Corda’ is a proprietary program for smart contracts and provides the ability to build apps on network

‘Concord Vault’ manages and records transactions in a Blockchain. It includes asset registry, trade registry and capacity to record cash balances. Transactions, although recorded are not made public. Only parties with legitimate need-to-know can see the data within an agreement.

It is well suited for governance, internal record keeping and regulatory. Smart contracts can resemble legal contracts making derivatives contracts, trade finance and commercial papers possible. By moving middle and back office functions to a secure cloud-based ledger, Concord will significantly streamline operations and lead to cost savings.

To sum up, it is unfair to simply state that one platform is superior to others, as they are all geared towards different objectives. In fact, some are inherently more suited to certain use cases depending on their capabilities and architecture.

[kunaljethwani]

Google, Facebook and other tech titans form 'Partnership on AI'