We Build Trust

Project Description

DealStreatAsia's website witnessed increasing traffic both from users and bots when we were approached to take over the site's maintenance and new development efforts. The site was infected with ad serving malware and the hosting costs were in excess of USD 1500 per month. Honey Badger Labs took over and implemented these measures to make the site secure, faster, more reliable while bringing down the hosting costs to a consistent USD 500 per month:

• Cleaned up malware from the source code and brought sources under version control.
• Applied security patches and implemented firewall controls.
• Added automated deployments and recovery procedures.
• Implemented caching to serve pages faster.
• Moved static media to CDN for more efficient delivery.
• Implemented S3 storage for all media for resiliency.
• Made search more efficient by offloading it from the database to ElasticSearch.
• Implemented HTTPS everywhere.
• Analyzed SQL queries and tuned them to reduce database load. Downsized the over provisioned database and web server to reduce costs and added multi zone replication for resilience.
• Offloaded image resizing load from the main server to a server-less micro-service implemented using Amazon Lambda.
• Reduced bot traffic by blocking bad bots and detecting crawler networks.
• Implemented paywall with Stripe integration with auto renewals. Paywall integrated using custom developed cache plugin to serve different versions to subscribers efficiently from the cache.
• Implemented new newsletter using MailChimp API to aggregate multiple feeds from the site.

Project Description

MyCloudForest wanted a custom built e-commerce site to sell balcony and landscape design products that offered customizations. They had a website built using Magento that did not offer the necessary customisations and it also had performance issues. We quickly built a site for MyCloudForest using a paid template on Django with the customisations they required. The site was redesigned with a custom design from our UX partner.

Project Description

Implemented a custom application for analysing performance of brands against their competition on digital platforms. This is a tool for analysts to collect around 100 data points (manually and using API) for each brand in the analysis and create a detailed performance report with actionable insights to improve the brand's performance. The project involved creation of customised interactive visualisations using D3JS.

Project Description

Pattern Effects Labs had built a prototype for predicting movements of stock indices using proprietary technology which had to be moved into production. The primary issues were around the efficiency of the algorithms which lead to scaling issues as well long testing and tuning cycles due to long turnaround times to check results from parameter modifications. Honey Badger Labs implemented vectorised versions of the algorithms using Python Pandas to cut down computation times for the core algorithm from 12 minutes to 40 milliseconds. The core components were separated out into smaller units communicating with each other over message queues. Custom visualisations were implemented for the user dashboard as well as for the validation interface to show index movement in real time along with predictions and recommendations. Technical analysis indications were also implemented using custom graphs in D3. The real time nature of the project required using a NodeJS to handle web socket channels to update user dashboards with low latency.

Project Description

Honey Badger Labs was the tech implementation partner in this project for analysing performance and fraud in the sales incentive program of a leading Indian FMCG company. The project involved analysing program data involving 70 million transactions.

The transaction details were provided as CSV dumps from different departments. The data sets were cleaned, lookup tables were implemented to generate correlation between the files from different departments. Additional computed fields were added to the schema for fraud analysis scripts that used K-means clustering. The data was loaded into a relational database for analysis using R Studio. The dashboard required creation of different types of aggregations for custom visualisations added by users. ElasticSearch and Kibana were chosen for this function. A massive ElasticSearch cluster was setup on Azure to handle real time aggregations from the 70 million records. After validation of the dashboard on Kibana, a custom dashboard was implemented in Django with the ElasticSearch cluster as the backend.

Project Description

Kamatan is a full basket fruit and vegetable supplier serving the HORECA (hotels, restaurants and cafes) segment. Honey Badger Labs designed and developed the initial ERP which handled procurements and sales. Kamatan has customised pricing contracts with each customer. The ERP was designed to generate invoices based on the pricing contracts and to analyse margins based in the plans. The sales and procurement data was used to build a margin analysis dashboard using Kibana and ElasticSearch. This dashboard enabled deep dive by customer, category and specific SKUs to identify pricing problems and customer's ordering patterns. Kamatan's team took over the development in a phased manner as they scaled up and built their own technical team.

Project Description

Vidki is an always on Android based video device used to connect two living rooms. Honey Badger Labs built a WebRTC based Android application and the necessary servers required for signalling and relaying media between the apps. A management interface for handling software updates and device configurations was also developed. The application featured light and motion detection algorithms to conserve bandwidth by switching frame rates since this was an always on video call. The devices would automatically lower the frame rates and bandwidth when both devices did not detect motion. Two years on, the application and the servers work with minimal maintenance for security updates.

Project Description

PropsAMC is a property management platform for end users, property managers and financial institutions. The platform is backed by legal, technical and financial data related to properties in major cities in India. Honey Badger Labs created the data schema and the necessary import tools to map PropsAMC's data (Excel files and Google Earth overlays in KML files) into a geo spatial database. User of the platform could locate their properties on Google maps and the platform pulled out details like zoning records and property prices trends around the property from the geo spatial database. The platform also implemented tools that allowed users to enter all financial transactions related to the property including purchase heads, loan details, maintenance costs and rental income. This data was used to calculate the XIRR on the property investment and generate comparisons with other financial instruments. PropsAMC also provides property related services like title verifications to their customers. Honey Badger Labs also developed a solution to take this business online using a combination of Zendesk for ticket workflow management and DropBox for secure handling of the property documents. PropsAMC took over the development in a phased manner and now develops and maintains the product with their own engineering team.

Project Description

Honey Badger Labs is a Freshworks development partner. When Freshdesk moved to a new UI and application platform, Honey Badger Labs implemented several applications that enhanced the core functionality of the ticketing platform. These includes extensions to manage automatic signature insertion, ticket field visibility and email behaviour based on the product or support groups. Honey Badger Labs also takes up custom development work for Freshdesk's customers to integrate their processes and platforms with Freshdesk.

Project Description

EagleOwl is a SaaS application developed by Honey Badger Labs for analysing restaurant operations. EagleOwl features analysis of sales, procurements and inventory to identify and eliminate inefficiencies, wastage and pilferage in restaurants. EagleOwl is already running successfully in some of the top breweries in Bangalore.