I believe SHAKEN/STIR was mandated only for calls originating from within the US, however, most of these fraud calls are originating over VOIP (or SIP) from outside the US.
Coinbase is a publicly listed company, with all its assets & liabilities shared in quarterly investor updates. They store all customers assets 1:1 (ie does not re-hypothecate any of customer assets)
1. When we started to specialize in a vertical and created the narrative and the product for that vertical. This is esp. true in domains with many legacy players.
2. For some reason or the other, we have always had delays in announcing our fundraising rounds: both Seed and Series A announcements were delayed by 3-6 months. The moment we announced, we got a step magnitude more inbound customers. Lesson here is to not be a perfectionist and be willing to share the company/product publicly earlier, so you can incorporate customer feedback early.
PS: If it helps provide context for (1) above, our domain is fraud prevention; and we focused on payment fraud for Fintechs/Crypto instead of going after generalized fraud prevention across all categories like ecommerce.
Handling ACH fraud is tricky. There is no decent dispute resolution mechanism in ACH compared to what Visa/MC offer for cards. So FinTechs and their backing bank are on the hook for customer reversals for up to 60 days. In the new FinTech revolution, this is one of the most misunderstood things. Many FinTech entrepreneurs think they can put up a nice UI and offer great interest rates only to get hit by huge fraud losses as soon as they launch. I run a fraud prevention startup for FinTechs - sardine.ai - and I hear of a new FinTech hit by fraud every week.
we are a stealth startup founded by former Paypal and Coinbase engineers. We have a device intelligence product that can detect accesses from proxies/VPNs without using any IP list. We can also detect the True OS that someone is using - useful to detect emulators and script kiddies. Happy to chat if of interest: info AT sardine.ai
Account recovery is a completely underserved problem today and many companies don't have a well thought through solution. Many rely on Knowledge Based Authentication questions that can be found easily by an attacker who has access to your email. To solve for cases where someone has lost all factors (forgot password, lost access to email, lost phone number), we need a brand new way of thinking about solutions. Hit me up if anyone's interested in brainstorming on the same!
Revolut | Head of crypto, Product owner - Fincrime, Backend dev, Data eng, ML eng | San Francisco, New York, London, Krakow, Vilinius | Onsite | Full time | www.revolut.com
REVOLUT - Fastest growing challenger bank in the world.
Our mission is to build the first truly global bank that provides zero fee FX trading across 29 currencies including 5 cryptocurrencies. We are rebuilding the banking experience: employee cards for businesses, small loans and stock trading, all within one app. We are planning to launch our product in a few months in Australia and US, and later in Singapore, Canada and Japan.
LOCATION: San Francisco, New York, London, Vilinius, Krakow
ROLES: Product Owner in Fincrime, Head of Crypto, Backend engineer (JAVA), Data engineer (Python), Devops engineer and many others
CULTURE: Everyone at Revolut is united behind the common mission of building a transparent and zero fee banking experience. We believe in meritocracy. We are a strong data-driven company and we believe in you can't improve what you can't measure. We are rebuilding banking from a technology and data first approach. Everyone at Revolut works hard, but not because anyone is asking them to, but because we are all passionate about our mission.
COMP & BENEFITS: salary + stock options vesting over 4 years. Additional stock options granted to high-performers every 6 months. Health care insurance + 401K matching.
TEAM: We have 700 employees of 60+ different nationalities worldwide across offices in London, Krakow, Moscow, St. Petersburg, New York and Vilinius. We are also opening a San Francisco office within the next few months. We still maintain the startup ethos with a flat hierarchy and many independent and autonomous teams that move quickly.
TECH STACK: Our tech stack is Java based and we are hosted on Google cloud with Kubernetes for cluster management. For data science and engineering, we use the latest open source tech including Apache Beam, Kafka, Airflow, TensorFlow, Catboost, to name a few.
ROLES in FINCRIME: I head Fincrime Risk (anti-fraud and anti-money laundering). I am hiring for the following roles in San Francisco:
> The opportunity here for a startup bank is to replace credit card spending in the USA. The potential earnings are huge if you can get consumers to spend on a debit card instead of a credit card (because the rake is higher).
I think the converse is true. Card issuers earn 1 percent points higher interchange rates on credit cards (2.60%) than debit cards (1.60%).
Yelp is co-sponsoring a new Data Science Contest. Use Yelp reviews in Boston to predict whether a business will have a poor Health Inspection score.
Participate in this contest to help in an important social cause. Cities have limited budgets and currently send Health Inspectors to restaurants on an ad-hoc basis. Using the Machine Learnt models built via the contest, they would be able to assign Health Inspectors to restaurants more efficiently. Past models have shown high accuracy (82%) since Yelpers often talk about the same things that health inspectors look for – cleanliness, ambience, methods of preparation. The contest started yesterday and will finish on June 23, 2015. Winners will win cash prizes from a pool of $5,000.
Two years, four highly competitive rounds, over $35,000 in cash prizes awarded and several hundred peer-reviewed papers later: the Yelp Dataset Challenge is doubling up. We are proud to announce our latest dataset that includes information about local businesses in 10 cities across 4 countries. This dataset contains 1.6M reviews and 500K tips by 366K users for 61K businesses along with rich attributes data (such as hours of operation, ambience, parking availability) for these businesses, social network information about the users, as well as aggregated check-ins over time for all these users. This treasure trove of local business data is waiting to be mined and we can’t wait to see you push the frontiers of data science research with our data. Submit your Data Science project by June 30, 2015 to win one of several prizes.
Gingerly (http://www.ginger.ly) is a Location-based Expense Tracking Android app that we've built on the side: think Mint meets Foursquare.
USP is our ambient sensing platform which shows you how much you've spent at a business, and your receipts for that place the moment you arrive there.
16K+ downloads and makes us about $50-100 per month (varies based on the month). Free app but Monetization is via in-app purchases (unlimited expense reports) and ads.
Expenses: $5 per month with a Digital Ocean droplet that hosts the website. Haven't had any maintenance issues and the app has been on auto-pilot mode for last 1 year.
With some repolish of the UI, this app has tremendous potential. A competing app with half the features but prettier UI ramped up to half-million downloads in 6 months.
If someone offers the right price, willing to sell it. Alternatively, also interested if someone wants to work with us on an iPhone app on an equity or rev-share basis.
So this would be similar to Impermium (now acquired by Google) meets Sift Science meets Cloud Flare - have you checked them out?
I have thought of building such a security as a service for websites for sometime. My PhD thesis was on building a reverse proxy server (kind of like nginx) to detect attackers (via machine learning) and starve them of CPU/network resources disproportionately (using admission control + WFQ) during a DDoS attack. After that I've spent a decade building systems to detect cyber security threats, click fraud, Twitter spammers, etc.
What do you think is lacking in existing security as a service offerings that prompted your post? Would love to chat more.