GIANT Stories at MongoDB

Building an AI-powered workplace ticketing system on MongoDB Atlas at Spoke

At every company, workplace teams are flooded with hundreds of questions on a daily basis. Support ticket platform, Spoke, aims to drastically simplify managing and responding to these time consuming queries. But Spoke goes beyond traditional ticketing with their friendly, AI-powered chatbot that gives workplace teams hours of time back as it automatically responds to questions on Slack, email, SMS, and web. And the more employees ask, the more Spoke learns.

Playback: Creating Operational Data Stores with Atlas and Stitch at Acxiom - MongoDB World 2018

Who wants to spend their time manipulating relational databases to be flexible about the data they store? And who would want to have a complicated and richly ticketed deployment process for each new client? Wouldn't it be great if MongoDB Atlas and Stitch could help make all that go away? That was the challenge that Acxiom's Scott Jones and Karen Wilson talked about in their MongoDB World 2018 presentation...

Playback: 3000 mongod nodes, 100 MongoDB clusters and automating with Ops Manager - MongoDB World 2018

Travel technology company Amadeus IT Group run some 100 MongoDB clusters with around 3000 mongod nodes. Some of those clusters have over 120 shards alone and at that scale, you need all the automation you can get your hands on. For Arkadiusz Borucki, Senior Service Reliability Engineer at Amadeus, that means using MongoDB Ops Manager as part of a completely automated environment. In his talk at MongoDB World 2018, he takes us through the range of technologies involved in that environment.

SynapseFI: Simplifying Banking For All With MongoDB Atlas

When your mission is to simplify banking and enable other companies to build the best-in-class financial product, you'll need to be fast, flexible and always ready to evolve. That's the business that SynapseFI is in, providing its customers with a wide variety of fast, easy and electronic financial services: sending and receiving payments, opening deposit accounts, white-labeled physical or virtual debit and credit cards, consumer and business loans, and much more. SynapseFI, with its fintech partners, has handled a total of $10 billion in transactions for close to 1.8 million end users to date.

Transforming Credit Management with Credisense and MongoDB

Credit Management is a grind -- clunky, time consuming and laden with risk.

It requires millions of dollars to capture consumer attention and nurture through the sales cycle. And then comes the arduous credit assessment, throwing a wrench into the promise of a seamless digital customer experience. In fact, up to 90% of bank new customer applications drop out due to slow onboarding 1.

Even once a deal closes, there is still plenty of work to do. The next hurdle is invoice collections, with default and delinquency rates averaging anywhere between 0.2-54.5% internationally 2.

In today’s digitally-driven market, consumers are demanding quicker turnaround times and instant approvals. This makes simplifying and streamlining the entire credit management process more critical than ever.

As MongoDB’s OEM business continues to rapidly expand, it’s a personal priority to work with organizations who are solving serious market needs with the most innovative technology. Credisense, our newest OEM partner, and their MongoDB- powered, full end-to-end origination and credit decisioning solution is a phenomenal example of this.

They’re already making splashes worldwide. For example: CTOS Data Systems (Malaysia’s largest credit reporting agency) is enabling banks, utilities, non-bank credit issuers, fintechs and lenders in the P2P lending space to make real time credit decisions using the Credisense platform, enabling things like instant loan approvals for credit cards and auto loans!

I had the opportunity to discuss the Credisense platform and the data technology behind it with Richard Brooks, Co-founder and Director.

Tell us a bit about yourself and the genesis of the company?

Our three co-founders have different, but complementary backgrounds. I have worked for bureau and data companies my entire career and been involved in the automation side quite extensively. Our second co-founder and CEO, Sean Hywood, is a software expert having built up several software companies over his career focusing on low-code technologies. We combined our knowledge with the technical expertise of our third co-founder and CTO, Waylon Turney-Mizen, with the vision of providing enterprise grade functionality to organizations of all sizes. The aim is to allow all businesses to make smarter decisions, faster.

For anyone that isn’t familiar with Credisense yet, could you describe why you set out to build this and the problem it’s solving?

Credit is a highly regulated, complex, often manual and costly process. McKinsey 3 rightly points out there are five key pressures on credit providers currently:

  1. Changing customer expectations, specifically digital and the customer experience
  2. Tighter regulatory controls such as AML/CFT and GDPR
  3. Data management, increasing reliance on clean data for analysis and decisions
  4. Market disruptor such as P2P lenders and digital banks
  5. Cost pressures driving down returns

There are some sobering stats that show how important these are, such as over $200 billion dollars 4 of regulatory fines in the US alone since the GFC, to the fact that traditional lenders have lost over 30 percent of personal loan market share 5 to agile financial technology companies. All these add up to some serious issue for business, some that even threaten their very existence. Our aim when creating Credisense was to tackle these issues, both by assisting traditional corporates to embrace this digital strategy, to providing this same technology and expertise to smaller businesses so they can compete and level the playing field.

How would you describe the platform and the unique advantages that Credisense gives its customers?

Our platform is born in the cloud and offers a “no-code” build capability allowing organizations to build out the functionality internally and grow the solution with their business. We have a unique graphical interface, and this coupled with the “no-code” technology allows business people -- not IT -- to build, own and manage the system.

The platform itself revolves around the decision and scoring engine which powers the advanced assessment and risk decisions for organizations.

Our MongoDB backend means we can confidently scale to handle millions of credit applications and still support real-time workflow and decision making in seconds.

How did you land on MongoDB to help you solve these challenges?

We needed a database to support a minimum of 100,000 transaction a day across a cloud platform. There are only a handful of NoSQL databases that can support the level of transaction with the ability to further scale if required. MongoDB ticked all the boxes. Add that to MongoDB’s great documentation security, tooling, support and APIs, and it made MongoDB the right choice for our development teams.

What advice would you give someone who is considering using MongoDB for their next project?

MongoDB offered us extensibility to be on-premises, which is something other cloud database platforms would not offer. It made sense to go with a database platform that offered both so that we could in turn offer this to our customers that require data to be held within their own environments for security reasons. Also, reach out and talk to MongoDB early in your process. The support they give you up front will help ensure you’re making the best decisions.

How are you securing MongoDB?

We utilize MongoDB Atlas for our Continuous Integration and Testing environment and will have a managed service offering. This is secured with an IP whitelist, secure password and SSL connection which was easy with Atlas and Atlas Professional.

We also have a customer-managed deployment secured out of the box behind a VPN that connects the app server to the MongoDB server. It also utilizes a strong username/password combination with minimum length and character requirements.

Through our OEM arrangement with MongoDB, we package MongoDB Enterprise as part of our product to ensure our customers have highly secure and enterprise-grade solutions.

Where have you deployed MongoDB? On-premises, in the cloud, via MongoDB Atlas? What tools are you using to deploy, monitor MongoDB?

All! The requirement for extensibility across platforms without any changes to the code was one of the key reasons for MongoDB selection.

MongoDB Atlas removes operational overhead and mitigates risk through automating many of the manual processes (configuring operating system, upgrades, backups and restores). This means we can focus on ensuring our customers have the robust platform they need to provide instant loan approvals.

We also have a production environment on-premises. Soon, we will introduce the use of MongoDB Cloud Manager for monitoring and alerts of on-premises production environments. With over 100 metrics and proactive alerting, we’ll be able to catch issues before they arise.



Fraud Detection at FICO with MongoDB and Microservices

FICO is more than just the FICO credit score. Founded in 1956, FICO also offers analytics applications for customer acquisition, service, and security, plus tools for decision management.

One of those applications is the Falcon Assurance Navigator (FAN), a fraud detection system that monitors purchasing and expenses through the full procurement to pay cycle. Consider an expense report: the entities involved include the reporter, the approver, the vendor, the department or business unit, the expense line items, and more. A single report has multiple line items, where each line may be broken into different expense codes, different budget sources, and so on. This translates into a complicated data model that can be nested 6 or 7 layers deep – a great match for MongoDB’s document model, but quite hard to represent in the tabular model of relational databases.

FICO FAN Fraud Detection Architecture FAN Architecture Overview

The fraud detection engine consists of a series of microservices that operate on transactions in queues that are persisted in MongoDB:

  • Each transaction arrives in a receiver service, which places it into a queue.
  • An attachment processor service checks for an attachment; if one exists, it sends it to an OCR service and stores the transaction enriched with the OCR data.
  • A context creator service analyzes it and associates it with any past transactions that are related to it.
  • A decision execution engine runs the rules that have been set up by the client and identifies violations.
  • One or more analytics engines review transactions and flag outliers.
  • Now decorated with a score, the transaction goes to a case manager service, which decides whether to create a case for human follow-up based on any identified issues.
  • At the same time, a notification manager passes updates on the processing of each transaction back to the client’s expense/procurement system.

To learn more, watch FICO’s presentation at MongoDB World 2018.

How Planable Uses Mongodb Atlas to Help Social Media Teams Move Their Creative Processes Forward

In 2015, Nicolae Gudumac was working at a social media agency managing hundreds of campaigns for clients. However, with each campaign requiring multiple rounds of review, he needed a way to streamline the disjointed feedback loop and bring everyone onto the same page. Along with his co-founders, Xenia Muntean and Vlad Calus, he began building a platform that would streamline this time-consuming process for social media managers, agencies, and their clients.

The team created Planable, a platform that simplifies planning, visualizing, and approving social media posts. The tool feels like a live mock-up of the social feed, making it easy for teams to collaborate and give real-time feedback in a familiar format.

Prioritizing Developer Velocity From the Beginning

As the newly minted CTO, Nicolae decided to use MongoDB’s document data model to help his team innovate quickly to stay ahead of the ever-changing social media landscape. “With social media content, requirements are constantly changing as platforms evolve and introduce new formats. Having a flexible data model has been a boon to our productivity.”

He also needed to build a foundation that could scale with them as the business grew. Since Planable is a collaboration tool that needs to keep users synced in real-time, Nicolae built the tool on top of Node.js and websockets. The team harnessed MongoDB’s oplog tailing functionality to send real-time updates to all connected users when relevant data changed and recently started to leverage MongoDB change streams to make this process more simple and scalable. To easily scale their app servers at peak times, Nicolae chose to run Planable on AWS container service.

Migrating to MongoDB Atlas

For managing their MongoDB deployment, the team started off using but were having issues with restoring backups and only had access to a very limited set of configuration options. Compose also charged a premium for upgrading their storage engine to WiredTiger. MongoDB Atlas, with its queryable backup snapshots, automated upgrades, and configuration flexibility — especially the ease at which clusters can be scaled horizontally — looked very appealing.


We needed to be able to provide our clients with a platform that is as reliable as we are. Atlas’ native cloud first and scale-out architecture aligned well with our increasing performance demands and usage growth.

Nicolae Gudumac, CTO, Planable

When Planable was accepted into the MongoDB Startup Accelerator, the timing was right to make the move to MongoDB Atlas. The team had customers all over the world and couldn’t afford any downtime with the migration. They used the Atlas Live Migration Service to move over their data from Compose with no downtime.

Currently, the team of 6 is split between engineering and business. With a small engineering team, they’re hyper-focused on product improvements and new features that will drive the business forward. Atlas features such as the Real-Time Performance panel and the Performance Advisor, which monitors clusters for slow queries and automatically suggests indexes to improve performance, allow the team to dedicate more of their attention to application improvements. According to Nicolae, “The Performance Advisor has made indexing the database and optimizing queries a no-brainer."

Queryable backups have also helped the team quickly address customer questions and as Nicolae recalls, “It’s saved us numerous times when someone accidentally dropped/updated a few documents and they needed to be restored. We’ve managed to quickly inspect and restore data in just a few clicks.”

The team is looking forward to the upcoming release of MongoDB Charts which is currently available in beta. “Charts will enable our marketing and business team to gain insights from our database, without resorting to sophisticated and expensive BI tools” says Nicolae.

Planable is bringing thousands onto their platform each month and are well on their way to becoming the default tool for social media collaboration.

MongoDB Atlas has helped this fast-growing startup focus on helping social media teams move their process forward and allows Nicolae to give valuable time back to his team. “There’s no doubt in my mind that moving to MongoDB Atlas has increased our team’s productivity.”

Atlas is the easiest and fastest way to get started with MongoDB. Deploy a free cluster in minutes.

Millions of Users and a Developer-Led Culture: How Blinkist Powers its Berlin Startup on MongoDB Atlas

Not unlike other startups, Blinkist grew its roots in a college dorm. Only, its creators didn’t know it at the time. It took years before the founders decided to build a business on their college study tricks. Blinkist condenses nonfiction books into pithy, but accessible 15-minute summaries which you can read or listen to via its app.

“It all started with four friends,” says Sebastian Schleicher, Director of Engineering at Blinkist. “After leaving university, they found jobs and built lifestyles that kept them fully occupied—but they were pretty frustrated because their packed schedules left them no time for reading and learning new things.”

Rather than resign themselves to a life without learning, they racked their brains as to how they could find a way to satisfy their craving for knowledge. They decided to revive their old study habits from university where they would write up key ideas from material that they’d read and then share it with each other. It didn’t take long for them to realise that they could build a business on this model of creating valuable easily accessible content to inspire people to keep learning. In 2012, Blinkist was born.

Six years later, the Berlin-based outfit has nearly 100 employees, but instead of writers and editors, they have Tea Masters and Content Ninjas. Blinkist has no formal hierarchical management structure, having replaced bosses with BOS, the Blinkist Operating System. The app has over five million users and, at its foundation, it has MongoDB Atlas, the fully managed service for MongoDB, running on AWS. But it didn’t always.

“In four years, we had a million users and 2,500 books,” says Schleicher. “We’d introduced audiobooks and seen them become the most important delivery channel. We tripled our revenue, doubled our team, moved into a larger, open-plan office, and even got a dog. Things were good.”

Running into trouble with 3rd party MongoDB as a Service

Then came an unwelcome plot twist. Blinkist had built its service on Compose, a third-party database as a service, based on MongoDB. MongoDB had been an obvious choice as the document model provided Blinkist with the flexibility needed to iterate quickly, but the team was too lean to spend time on infrastructure management

In 2016, Compose unexpectedly decided to change the architecture of its database, creating major obstacles for Blinkist as they would become locked-in to an old version of MongoDB. “They left us alone,” says Schleicher. “They said, ‘Here’s a tool, migrate your data.’ I asked if they’d help. No dice. I offered them money. Not interested, no support. After being a customer for all those years? I said goodbye.”

After years of issues, it became clear last year that Blinkist would need to leave Compose, which meant choosing a new database provider. “We looked at migrating to MySQL, we were that desperate. That would have meant freezing development and concentrating on the move ourselves. On a live service. It was bleak.”

Discovering MongoDB Atlas

By this time, MongoDB’s managed cloud Atlas service was well established and seemed to be the logical solution. “We downloaded MongoDB’s free mongomirror service to make the transition,” says Schleicher, “but we hit a brick wall. Compose had locked us into a very old version of the database and who knows what else, and we couldn’t work it out.”

At that point, Schleicher made a call to MongoDB. MongoDB didn’t say, ‘Do it yourself.’ Instead, they sent their own data ninja—or, in more conventional, business-card wording, a principal consulting engineer. “It was the easiest thing in the world,” Schleicher remembers. “In one day, he implemented four feature requests, got the migration done and our databases were in live sync. Such a great experience.”

Now that Blinkist is on Atlas, Schleicher feels like they have a very solid base for the future. “Performance is terrific. Our mobile app developers accidentally coded in a distributed denial of service attack on our own systems. Every day at midnight, in each time zone, our mobile apps all simultaneously sync. This pushes the requests load up from a normal peak of 7,500 requests a minute to 40,000 continuous. That would have slaughtered the old system, with real business impacts — killing sign-ups and user interactions. This time, nobody noticed anything was wrong."


Right now it feels like we have a big tech advantage. With MongoDB Atlas and AWS, we’re on the shoulders of people who can scale the world. I know for the foreseeable future I have partners I can really rely on.

Sebastian Schleicher, Director of Engineering, Blinkist

Schleicher adds: “We’re building our future through microarchitecture with all the frills. Developers know they don’t have to worry about what’s going on behind the API in MongoDB. It just works. We’re free to look at data analytics and AI—whatever techniques and tools we believe will help us grow—and not spend all our time maintaining a monolithic slab of code.”

With Blinkist’s global ambitions, scaling isn’t just a technical challenge; it tests company culture—no matter how modern—to the limits. MongoDB’s own customer-focused culture, it turns out, is proving as compatible as MongoDB’s data platform.

“Talking to MongoDB isn’t like being exposed to relentless sales pressure. It’s cooperative, it’s reassuring. There are lots of good technical people on tap. It’s holistic, no silos, whatever it takes to help us.”

This partnership is helping make Blinkist a great place to be a developer.

“A new colleague we hired last year told me we’ve created an island of happiness for engineers. Once you have an understanding of the business needs and vision, you get to drive your own projects. We believe in super transparency. Everyone is empowered.”

“Oh, and did I mention we have a dog?”

Atlas is the easiest and fastest way to get started with MongoDB. Deploy a free cluster in minutes.

Leading digital cryptocurrency exchange cuts developer time by two-thirds and overcomes scaling challenges with MongoDB Atlas

Cryptocurrency investing is a wild ride. And while many have contemplated the lucrative enterprise of building an exchange, few have the technical know-how, robust engineering, and nerves of steel to succeed. Discidium Internet Labs decided it qualified and launched Koinex, a multi-cryptocurrency trading platform, in India in August 2017. By the end of the year, it was the largest digital asset exchange by volume in the country. Turns to MongoDB to Build a Single Customer View

Transforms customer experience, fights fraud, and meets the demands of the GDPR with MongoDB Atlas and Apache Kafka running on AWS