Lambert Labs

Boost simplifies stock ordering for small businesses in Africa. Businesses can order inventory via Boost’s straightforward WhatsApp interface and have it delivered right to their door in a timely fashion. With Boost, a local shopkeeper or restauranteur can easily order whatever produce they need whilst keeping their business open for custom. Boost currently operates in Ghana, Nigeria and South Africa and is expanding rapidly across Africa.

Boost’s platform is predominantly Python-based and hosted on Amazon Web Services (AWS) using AWS Elastic Beanstalk, with data persistence provided by Amazon RDS and Amazon DocumentDB. Vonage powers Boost’s implementation of the WhatsApp Business API. The platform provides web applications  for customers and admins alike, as well as an offline-capable web app for purchasing stock, with data persistence provided by IndexedDB.

Why Lambert Labs?

Boost reached out to Lambert Labs for help across their technology stack, from cloud computing with AWS to Flask API endpoints to UI improvements in the frontend.

At Lambert Labs we are AWS Select Partners, are Python experts and have reasonable frontend experience, so we were well placed to help the team at Boost!

What did we do?

This is a snapshot of the work we have completed for Boost thus far:

– Setup an event-streaming system in AWS using Apache Kafka and Amazon EC2. This reliably delivers customers’ messages to a Natural Language Processing (NLP) engine, reducing timeouts and enhancing the overall user experience.

– Programmed an offline app for purchasing stock. Data persistence is provided by IndexedDB.

– Set up a Celery worker cluster in AWS ECS to pass off  a key step in the online ordering process to background workers, and enable other scheduled tasks.

– Deployed a secure staging environment on AWS Elastic Beanstalk, with scheduled database sync jobs executed using AWS Lambda and AWS Step Functions.

Who are Lawyers on Demand?

Lawyers on Demand provides flexible legal solutions to a diverse range of organisations across the public and private sectors. It uses a freelance model to provide a variety of services including managed services, secondments, legal operations and technology.

The Challenge

Lawyers on Demand wanted to make improvements to its website in order to drive an increase in the number of potential clients getting in touch to acquire legal expertise.

Why Lambert Labs?

Lawyers on Demand chose to work with Lambert Labs because of our expertise with Python, Django, Wagtail, PostgreSQL, Kubernetes and Azure.

What We Did

• Worked closely with the Chief Information Officer and Head of Software Development to ensure the security and robustness of the website.
• Worked closely with the Commercial Director and Marketing Team to translate marketing requirements into technical requirements for the website.
• Made continuous improvements to the Wagtail administration panel to enable the Marketing Team to produce more customised content at speed.
• Added the ability for the Marketing Team to embed Vimeo videos, HubSpot forms and podcasts into the website.
• Improved the Meet the Team page so that consumers of the website can now easily find team members based upon their location or job function.
• Improved a continuous integration pipeline on Azure DevOps so that it now includes linting of Python, TypeScript and CSS.
• Improved a continuous deployment pipeline on Azure DevOps so that it now includes builds of JavaScript and CSS bundes.
• Managed tens of website feature releases through a multi-region continuous deployment pipeline using Azure DevOps, Azure Front Door and Kubernetes.

Impact

• The website conversion rate has increased significantly.
• The Marketing Team is now able to produce more customised content via the Wagtail administration panel.
• New website features can be deployed through the continuous integration and deployment pipelines more quickly and reliably.

Who are Sciurus Analytics?

Sciurus Analytics is an established pensions and actuarial software provider. Their innovative and powerful solution enables pension providers, fund managers, and life insurers to efficiently build and share complex financial modelling and analytics. 

The company has a unique blend of industry, mathematical, and software development expertise. Their software consists of Financial Canvas, a desktop application, which links to Financial Canvas Studio, a SaaS offering. Their products are used by hundreds of finance professionals at large pension schemes and multinational companies with over 800 schemes and £100bn of assets under analysis.

Why Lambert Labs?

Lambert Labs were chosen because of our proficiency in cloud computing, specifically our expertise in deploying end-to-end production workloads to Amazon Web Services (AWS) according to industry best practices, e.g. the AWS Well-Architected Framework. Our AWS consulting expertise is backed up by our status as AWS Select Partners.

Sciurus Analytics also enlisted our services because of our skills and experience developing web applications in Python and ReactJS.

What we did

• Production deployment of a React app with rich interactive visualisations over HTTPS using CloudFront and S3 that has successfully scaled across multiple clients
• Production deployment of a backend Flask API over HTTPS using a load-balanced Elastic Beanstalk application, connected to a MongoDB database (AWS Marketplace; Bitnami).
• Security hardening of the API deployment via network and firewall configurations to pass external penetration testing
• Production deployment of file servers on Windows Server EC2 instances and using AWS FSx
• Production deployment of CPU-intensive MATLAB workloads on Windows Server EC2 instances triggered via Systems Manager
• Development of new features and bug fixes using Python (Flask; PyMongo; custom MongoDB ORM), React (Redux; D3; Bootstrap)

Who are Caribou Data?

Caribou Data provides insights into how consumers in emerging markets interact with digital products and services on a day-to-day basis. The business is built upon a core principle of handling data ethically; one of Caribou Data’s key tenets is protecting user privacy and anonymity.

Why Lambert Labs?

Caribou Data came to Lambert Labs because of our experience and expertise in a number of areas: writing robust, production-quality application code in Python; strong knowledge and familiarity with the Python ecosystem; deploying web apps to the cloud; extraction, manipulation, and analysis of data.

Further, as a software development agency taking pride in our work, we ensure clients’ projects are setup and managed using processes enabling scalability in terms of the number of developers working on a project and agility in terms of changing business requirements and industry trends. Notably, we achieve the latter by deploying code iteratively and frequently using automated tools, including continuous integration and continuous delivery/deployment (CI/CD).

What we did – technical implementation

• Configured CI/CD for several different microservices using CircleCI.
• Deployed three HTTPS-enabled Django apps to the web using AWS Elastic Beanstalk, AWS Elastic Container Registry, and PostgreSQL AWS Relational Database Service.
• Created and deployed Python packages to the cloud using Github.
• Added API functionality to Django web apps via configuration of the Django Rest Framework.

Extraction, transformation, and loading of data (ETL):

• Created an event-driven ETL pipeline using AWS Lambda (in conjunction with Serverless), including data quality and pre-processing.
• Created ETL jobs using PySpark on AWS Glue.
• Integrated the pipeline with AWS S3 and Redshift.

What is Bongo’s Bingo?

Bongo’s Bingo is a bingo-based club night founded in Liverpool in 2015. It has since exploded in popularity to the extent that over 500 Bongo’s Bingo events were held UK-wide and worldwide in 2019, including nights in Amsterdam, Paris, Ibiza and Sydney. Events consist of several rounds of bingo mixed with rave intervals, dance-offs and catwalks. They are basically bingo with jugs of drinks, wild dancing to retro pop music and celebrity appearances.

Why Lambert Labs?

Bongo’s Bingo came to Lambert Labs for help with maintaining and improving their website and the backend of their mobile app because of our expertise with Python and our knowledge of DevOps/AWS.

We quickly setup a scalable website/API deployment using AWS Elastic Beanstalk. The application has over 300,000 users and the website sold more than 170,000 tickets in 2019.

What we did

CI/CD and DevOps:

• Deployed a Django web application using AWS Elastic Beanstalk, Amazon Elastic Container Registry, PostgreSQL on Amazon RDS and AWS Certificate Manager.
• Setup continuous integration and deployment for all microservices using CircleCI.
• Setup robust database backup and restoration processes.

Python application code:

• Worked on an existing Django web application.
• Maintained and improved an existing Stripe integration in order to streamline the customer payment process.
• Added API functionality to the Django application via the Django Rest Framework.

Testing and quality control:

• Built an automated browser test suite using Python and Selenium. The test suite runs against both the local, staging and hotfix environments to ensure no bugs creep into production.

Logging and monitoring:

• Configured dashboards in Amazon CloudWatch to monitor application and database servers in production and staging environments.
• Setup CloudWatch alarms and uptime alerts using AWS Simple Notification Service.

Who are Sentient Machines?

Sentient Machines is an award-winning startup that uses proprietary artificial intelligence to analyse customer and employee interactions. It provides a summary of the results as well as actionable insights back to their clients via a web platform. The web platform flags emerging issues, non-compliance, and agent performance challenges, including otherwise difficult to identify dimensions such as empathy and listening skills. Users can also view audio recordings of conversations as well as textual records of emails and chats that are annotated by the AI.

The web platform is designed using React JS on the frontend and Scala’s Play framework on the backend. The service uses a Mongo database and its technologies are mainly hosted on Google Cloud.

Why Lambert Labs?

Sentient Machines reached out to Lambert Labs for help with improving the user experience of their web platform because of our expertise with various JavaScript frameworks and knowledge of MongoDB.

What did we do?

• We rewrote all Scala API endpoints to allow users to filter data by both the day of the week and the time of day. We redesigned an open source React JS date picker to make use of these options. To ensure that the database was being filtered in the most efficient way, we performance tested our Mongo queries, eliminated existing bottlenecks in the JavaScript codebase, and also helped to upgrade the database version.
• We allowed users to have conversations on the web platform around different audio recordings by designing a messaging system and implemented Mailgun’s API to notify users by email whenever they have been messaged.

Who are Yagro?

Yagro is an agricultural technology business with a mission to connect agriculture and create a more sustainable farming sector. It offers a variety of commercial management tools designed to help farming become more profitable.  Products include YAGRO Analytics, which enables farms to make better business decisions, YAGRO Insurance, which offers insurance for farms and YAGRO Marketplace, which enables farms to order from suppliers in a marketplace with live prices.

The Challenge

Yagro needed to build a trading platform which would allow agricultural companies in South America to buy and sell different products, including fertilisers.

Why Lambert Labs?

Yagro chose to work with Lambert Labs because of our experience of developing software applications using Python and our expertise working with Flask, Django and PostgreSQL.

What We Did

• Created the ability for agricultural companies to make bids for, offers for and trade products, including fertilisers.
• Collaborated with the Head of Engineering to architect and develop a Flask-based API microservice to fit in with Yagro’s existing microservices architecture.
• Designed an appropriate PostgreSQL database schema and used SQLAlchemy combined with Flask-Marshmallow to manage data inside the Flask application.
• Used Alembic to manage database migrations for the Flask application.
• Implemented API endpoints to allow the users of the frontend to create, update and delete bids and offers and make trades.
• Integrated the Flask microservice with a pre-existing Django-based internal API where appropriate.
• Implemented integration tests using pytest.
• Worked closely with the Head of Product to design and develop the front end of the trading platform using Vue.js

Impact

The trading platform has now been released and is being used in Brazil.

When a transnational company such as Accenture want expertise in a particular area they often use suppliers with niche skillsets. Accenture recently turned to Lambert Labs for help with prototyping improvements to a website built with the Django Wagtail CMS. In this case, the end client was the NHS.

We built a prototype to demonstrate how Draftail can be used to improve the content authoring process and how schema.org metadata can be used to improve SEO.

Who are Aerial Empire?

Aerial Empire specialise in producing video footage of outdoor spaces. As part of their business, they create and deliver 360-degree timelapse footage of construction sites, giving companies an accessible overview of progress on the ground. They also produce artistic drone films for clients such as tourism companies and property agents.

Why Lambert Labs?

Aerial Empire needed help in expanding their timelapse video management system, the “Time-Lapse Portal”, across all aspects. This Portal runs as a Python/Flask app hosted on Digital Ocean. As highly-skilled Flask developers, we were well suited to the task. In addition, video production is managed by a Celery worker, which in turn is one of many containers orchestrated by Kubernetes in the cloud. Both of these technologies are ones with which members of our team have had extensive experience.

What we did

• Scaled up resources on the compute instance responsible for video production. This meant we could render longer and higher-resolution timelapses without worrying about out-of-memory crashes.
• Implemented a custom Python script to scale up/down cloud resources as needed to help reduce business costs.
• Increased the video editor capabilities in the Portal, so that Aerial Empire could customize more attributes like video resolution directly from the frontend.
• Improved the self-render functionality of the Portal, which periodically re-renders timelapse videos to include the latest images captured by on-site cameras.
• Helped implement a UI upgrade of the Portal.
• Streamlined the video rendering algorithm by optimizing our usage of FFmpeg encoding features.

What is OpenClassrooms?

OpenClassrooms is an online education platform which provides courses in technology, entrepreneurship and digital skills to more than 2.5 million users worldwide. The courses enable users to learn online by reading, watching videos, doing projects and participating in mentoring sessions.

The Challenge

OpenClassrooms wanted to design and build a 12-month bachelor-level Python diploma which would prepare students for entry-level jobs in software development using Python.

Why Lambert Labs?

OpenClassrooms chose to work with Lambert Labs because of our ability to provide a committee of Python experts with a range of skills and experience to help with the design and creation of their Python diploma path.

What We Did

• Collaborated with other Python experts and a Senior Instructional Designer at OpenClassrooms to design the curriculum for the Python diploma path. This included outlining courses and projects for students to complete as part of the diploma.
• Produced detailed plans for student projects, including project briefs and mentor guides. These projects have titles including ‘Use Python Basics for Market Analysis’, ‘Develop a Software Program Using Python’, ‘Develop a Web Application Using Django’ and ‘Scale a Django Application Using Modular Architecture’.
• Authored a course titled Set Up a Python Environment. The course teaches students how to manage Python packages with pip, how to create and manage Python virtual environments with venv and how to set up a project in an IDE using PyCharm.
• Reviewed other Python experts’ project and course outlines and content in order to ensure high quality.

Impact

• The OpenClassrooms Python diploma is now available to students globally.
• The student projects that we planned, designed and produced are available as part of this diploma.
• The Set Up a Python Environment course that we wrote has been published.

Cisco Meraki is a San Francisco-based global leader in cloud controlled WiFi, routing and security solutions. Their products include security cameras and sensors, wireless networking, switching, mobile device management and wide area networking.

Why Lambert Labs?

Cisco Meraki chose to work with Lambert Labs because of our expertise in using Python data cleaning processes in order to transform unstructured data into structured data. 

What we did

The team at Lambert Labs worked closely with the Cisco Meraki sales team to transform an unstructured, messy dataset of sales leads into a structured clean dataset of sales targets for a new product launch. 

Our implementation included heavy use of vanilla Python, including extensive use of regex for text cleaning and manipulation. 

To ensure the easy maintainability of our codebase we used our standard set of tools for static code analysis, including 3rd-party linting packages such as flake8, flake8-bugbear, flake8-builtins, flake8-commas, flake8-comprehensions, flake8-print, flake8-import-order and flake8-quotes. We use pytest for our unit tests.

Let’s Do This (Y Combinator winter 2018) are a fast-growing startup aiming to be the Airbnb of the sports endurance world. We carried out of one of their key data analytics objectives, replicating NoSQL (MongoDB) data into a relational database (Amazon Redshift) in a format designed for maximum ease of querying for staff across their business.

To speed up the process and respect budgeting requirements, we used a third party data integration provider Stitch, whilst also writing our own SQL in order to tailor the results to the exact specifications of the client.

As with many technology companies today, part of Let’s Do This’ business model relies on gathering data from a range of different sources. We used our software development expertise to create a data aggregation and cleaning pipeline that is efficient, easily extensible / maintainable and scalable, leveraging modern industry best practices along the way.

Fundment approached us with a quality assurance project to tackle. They wanted automated browser testing for their site so they could develop and improve their product, safe in the knowledge that any bugs would not find their way into production.

To tackle this task we used a selection of tools to build a testing suite that could be used to rigorously test the functionality of the site. Pytest and Selenium were used to replicate a user interacting with the application, doing everything from completing the sign up process to making a bank transfer.

Being able to test the functionality of the site via an automated browser session is powerful, but when you combine it with a tool such as BrowserStack it really comes into its own. Browserstack allowed us to undertake cross-OS and cross-browser testing, meaning that the team at Fundment could be confident that their site was working as expected, regardless of whether the customer was using Internet Explorer on Windows 7 or Firefox on MacOS.

These tests are now running on a remote server every night, taking several hours to complete due to their sheer scope. When completed they provide a custom health-check report via email, meaning the development team at Fundment can rest easy knowing their site is being tested to the most rigorous of standards.

Who are CoGo?

CoGo (Connecting Good) is an innovative social impact startup giving consumers the opportunity to spend their money more ethically. Their mobile app enables users to connect with businesses sharing similar values on social issues, including climate change, fair trade, and living wages. App users can find like-minded businesses and are encouraged to spend their money more ethically.

Why Lambert Labs?

CoGo reached out to Lambert Labs because of our expertise in automated testing and experience creating test suites enabling efficient and rigorous Quality Assurance (QA).

In particular, Lambert Labs’ in-depth knowledge of Python and testing libraries such as pytest and Selenium made us a good fit for CoGo’s QA objectives.

What we did – technical implementation

• Setup a local automated testing environment using Python, pytest, Appium, and Selenium
• Configured automated testing for Android devices (emulator and physical) via Java Development Kit and Android Studio
• Configured automated testing for iOS devices (emulator and physical) via Xcode
• Integrated the automated testing suite with the CoGo mobile app, which is built using React Native
• Wrote tests for Android and iOS covering core user actions including login, logout, signup and various other user journeys
• Adapted the pytest command line interface to improve developer productivity
• Created logging of failed tests (log messages and app screenshots) to facilitate analysis of the app’s functionality
• Updated tests regularly in line with new versions of the app

David Morgan Education develops products to help children to read and spell. They came to us asking if we could rebuild and revamp the learning management system (LMS) of their existing product, Easyread.

Of course, aims and ambitions are always likely to evolve as the project matures. We quickly realised that, in order to make the LMS reusable for other projects in their pipeline, we should be building a framework, rather than a standalone site. In fact it eventually came to be that we launched the LMS with one of these new projects, Peasyread, and later transferred it to the Easyread product.

In rebuilding their site we decided to replace their existing PHP backend with a Django based one. This meant that our framework was structured much in the way as a Django Package is. After careful planning of database structure we decided to favour the rigidly defined schemas of SQL over NoSQL databases, and opted for PostgreSQL due to its modern features and fast performance. We chose Docker to quickly create reusable Linux-based images from which to run the site and hosted these server side with Kubernetes, which allows for resilient and scalable management of Docker containers. We hosted our servers in the cloud using Google Kubernetes Engine on Google Cloud Platform. Webpack and Google Storage were used to package and host JavaScript, CSS and other assets.

Who are Reio?

Reio provides teams and organisations with tools to help streamline their Human Resources processes and systems. Their mission is to allow companies to run more effectively by reducing the amount of time spent on tasks such as calculating payrolls or managing holiday bookings.

Why Lambert Labs?

Reio were in the process of moving to a React frontend for their main web app, and were looking for someone to assist in adapting the Django backend for this purpose. Being experienced developers in Django and Python, Lambert Labs were well suited to this task. In addition, code cleanliness and maintainability is a key priority of Lambert Labs, so Reio could rest assured that they can build on our work in the future.

What we did

• Integrated parts of the Django backend with React and Redux on the frontend, using the Django REST Framework.
• Identified, diagnosed and fixed various bugs related to the existing React/Django integration.
• Helped upgrade the existing payment processor to Stripe by building out base methods in Django.
• Provided directions on future Stripe development.

Who are Step Function?

Step Function provide training in Data Science, Data Analysis and Python to individuals and corporates. Courses range from 1-day introductions for beginners to 4-week intensive programmes. The courses are either delivered as evening classes and weekend courses (for people already in full time jobs), or as more intensive courses for companies wanting to train their staff.

Why Lambert Labs?

Step Function chose to partner with Lambert Labs because of our deep knowledge of Python and our considerable experience with delivering teaching/training in both formal and informal environments.

What we did

Lambert Labs delivered Introduction to Python courses to both individuals (evening classes) and corporates (including apprentices on an apprenticeship scheme run by The Challenge). Topics covered included:

• Data types: ints, floats, Booleans and more.
• Data structures: strings, lists, dictionaries, tuple, sets and more.
• Control flow: conditional statements and loops.
• Built-in and user-defined functions.
• Classes and object-oriented programming.
• Exception handling.
• Using Python to make HTTP requests.
• Using Python to interact with APIs. 

Today, the Cloud and DevOps are the key driving forces and building blocks behind many companies’ technology infrastructure.

Enforcd (a RegTech startup with a global intelligence platform for financial services firms) tasked us with migrating some of their existing services to Amazon Web Services (AWS) in order to move from physical hardware-based computing / networking resources to virtual resources in the Cloud.

Lambert Labs set up a Virtual Private Cloud (VPC) to replicate Enforcd’s existing infrastructure (which made use of subnets, multiple IPv4 / IPv6 addresses, and LXC containers). Given Lambert Labs’ familiarity and expertise with AWS, we were able to accomplish the task in a timely fashion with minimum fuss.

Post-migration, these services can now harness the power, security, and flexibility of a leading cloud provider, as well as lend themselves easily to integration with DevOps methodologies in the future.

Enforcd also reached out to Lambert Labs for technology consultancy when they were in the process of being sold. Lambert Labs advised on technical aspects of the sale, including:

• Compiling technology architecture summaries for prospective buyers.
• Managing AWS IAM permissions throughout the sales process and at the point of sale.

Who are High Voltage Analytics?

High Voltage Analytics is a technology consultancy operating in the e-Commerce space. They provide the necessary resources so their clients can fully leverage the latest technology to scale successfully and build quality infrastructure using the modern industry techniques and best practices.

Why Lambert Labs?

High Voltage Analytics decided to collaborate with Lambert Labs because of our skills and experience using, developing and architecting extract, transform, load (ETL) solutions for business intelligence (BI) workloads.

High Voltage Analytics chose us to build an end-to-end BI prototype for their client, a major player in the E-Scooters industry, because of our ability to quickly understand the problem and outline appropriate solutions given the project’s objectives and time/budget constraints. Crucial to this was our understanding of key trade-offs, e.g. using custom vs off the shelf solutions.

What we did – technical implementation

• Setup a Stitch integration to copy data from multiple Shopify accounts into Amazon Redshift on a regular schedule.
• Wrote custom SQL to generate business-critical KPIs and metrics for each Shopify account as specified by the end client.
• Wrote custom SQL to aggregate data at the account level and produce KPIs and metrics for the entire business.
• Integrated the resulting data with a frontend BI visualisation tool, Grow BI, so that end users could view and analyse the data via the web.
• Identified the limitations of the prototype and the ways that these could be addressed in the future: greater automation, CI/CD, templated data transformations, integration with AWS services.

What we did – non-technical implementation

• Organised regular calls for progress updates.
• Organised calls with external vendors ensuring participation and feedback from relevant stakeholders.
• Produced documentation giving a high level overview of the prototype architecture.
• Produced marketing / sales material for future projects.

Yousoft recently approached us with the task of building a program which would scan their clients mobile application for bugs. As a Python development agency experienced with testing browsers using Selenium, we knew that Python is ideal for writing automated tests and, together with Selenium’s extension Appium, has the power to run tests on mobile devices.

Using these tools, we set to work developing a testing suite which can run against an Android app emulated on a Windows 10 machine. Our code was hosted on GitHub, and we used Conda to manage our Python environments. We also trained the Yousoft development team so that they could manage and maintain the testing suite on an ongoing basis. 

At Lambert Labs we always try to follow best development practices. We used CircleCI to run tests which, as a bare minimum, check the Python code for bugs and styling issues before any kind of update is made to the code base. By integrating CircleCI with GitHub, we ensured that the testing suite was always maintained to the highest standards of programming.

After 3 months of consultation with Yousoft, Lambert Labs produced a testing suite which can perform any user action desired on the app. This includes everything from registering a user, all the way through to the most niche user journeys. Thanks to this piece of software, Yousoft can rest easy that all of the critical user flows run smoothly on each new version of the app they produce, and that any bugs in these flows will be found and quickly dealt with before spreading into future production builds.

Who are Saffronblue.ai

Saffronblue.ai is an AI based software house that builds human-to-machine learning systems across industries. 

Why Lambert Labs?

Saffronblue.ai chose to work with Lambert Labs because of our in-depth knowledge of Python, our experience with hosting products on AWS and our robust workflow processes.  

The product

Lambert Labs designed, built and now manage the hosting for one of Saffronblue.ai’s products in the finance space, a B2B API. Users can upload data to the API, which connects to the Saffronblue.ai artificial intelligence engine, processes the data and returns results.

What we did – technical implementation

• Deployed the API to Amazon API Gateway using the Serverless framework.
• Wrote AWS Lambda functions for each API endpoint using Python.
• Deployed the Saffronblue.ai artificial intelligence engine using AWS Batch, with Docker images stored in Amazon Elastic Container Registry.
• Configured staging and production databases using MongoDB Atlas.
• Configured API documentation and other static websites using Amazon CloudFront and Amazon S3.
• Deployed a Flask application using AWS Elastic Beanstalk.
• Setup logging and monitoring using Amazon CloudWatch.
• Setup continuous integration and deployment for all repositories using CircleCI.
• Configured local, staging and production environments.

What we did – non-technical implementation

• Helped to setup and use a robust workflow on Jira.
• Used Slack for team communication.
• Participated in daily standup calls using Hangouts.

Who are Fundment?

Fundment are a fintech company for financial advisers and their clients. Using proprietary techonology, their platform provides an enhanced digital investment experience for its users, allowing them to automate, streamline, and modernise their investment activities and administration.

Why Lambert Labs?

Fundment came to Lambert Labs having already worked with us on an automated testing project for one of their web apps. Lambert Labs' deep and wide-ranging knowledge of the Python language and ecosystem, together with our strong expertise in cloud computing (Google Cloud Platform), made us a natural choice.

What did we do?

The main focus of our work was a large backend API powering several of the company's applications and systems.

Tasked with a complete rewrite of the original Hug API, we created over a hundred endpoints from scratch using FastAPI (adding both new and replacing existing functionality). The new API was RESTful and followed the OpenAPI specification.

Technical highlights of the new API were:

• Rigourous, well structured and documented schemas using Pydantic
• Efficient and Pythonic database querying using SQLAlchemy
• Modular, well designed, secure, and easy to maintain code via FastAPI features, e.g. routers, dependency injection
• Automatically generated and interactive Swagger API documentation
• Unit, integration (end to end API endpoint tests), and functional (database functions) testing (Pytest) with high test coverage (over 500 tests)
• Authentication - implemented an OAuth2 (Google and Microsoft) login
• Python package development - abstracted core functionality into a standalone library used across the company, hosted privately in the cloud; we were reponsible for new features, maintenance, and releases

We integrated smoothly with the Fundment frontend team consuming the API (issue triaging, delivering feature requests, bug fixes) and provided onboarding and technical support to other API consumers, e.g. Fundment's clients.