Our Work
With Amytics knowledge & experience, Accelerate your strategic decisions and drive success with fast and accurate insights
Data Science & Machine Learning
Text analytics and NLP to predict financial indicators from annual business reports
-
Financial management firm client with core requirements to reduce the manual analysis of annual financial reports and get automated insights to predict future trends.
- Processed text and developed an NLP system to analyse and extract financial indicators from annual reports to automate the process of analyzing which actually takes hours or days manually.
-
Implemented classical and advanced deep learning NLP techniques to provide an end-to-end data science life-cycle solution from data collection, preprocessing, eda, analysis ,modelling, evaluation, deployment, scheduling and monitoring.
-
Simple client friendly API based solution to easily upload and analyze documents automatically, resulting in overall business sentiments distribution over the reports.
Big data analytics to match unified customer identity across large customer base
-
Client is a one of the major information security company looking for a solution to match unified customer identity across their large consumer base of ~12 million from different industries such as e-commerce, retail, digital services and CRM.
-
Built a machine learning process pipeline to extract data from different data sources, run preprocessing on text, perform our matching mechanism and load processed output data into delta lakes.
-
We came up with a unified solution leveraging the use of big data, NLP and text mining techniques with PySpark on the Databricks platform. Successfully reduced the processing time by 30% by using optimal data science techniques with spark.
Portfolio management framework to automatically predict stock performance
-
Worked collaboratively with a major
asset manager to develop a process framework to automatically manage portfolios & predict stock performance.
-
Developed machine learning models to perform the predictions accurately with proper evaluation/validation strategies to improve the desired metrics.
-
Built data pipeline to extract data, process and perform predictions as per scheduled timeline and leveraged fully managed and monitored workflows using airflow and logging dashboards to keep track of performance metrics and visualization.
Detecting grades for industrial products using vision based analytics
-
Industrial manufacturer client with a usecase to automatically detect product grades from industrial machinery in real time.
-
Developed a computer vision based solution to perform classification on different product grades running on
industrial machinery.
-
Used model compression techniques to deploy solution using IOT technology, resulting in reduced inference time to perform low-latency predictions on
micro-controllers.
Big Data Engineering
Customer Data Platform
-
Customer 360 degree platform to help the business make quicker decisions & to be marketable.
-
We have helped our customer to create a modern scalable data pipeline which assembles data from various sources and various formats.
-
Employed Open-Source ETL tools to create reliable pipelines capturing data from databases, social platforms, CRMs and Marketing.
-
We have achieved an organised data platform with an organic as well as transformed data to help the business with the near real time business insights.
Product Search Engine
-
One of the growing manufacturing firm was willing to develop a search engine for its customer to create a market place.
-
Stream the data from various 3rd party vendors and extract records from the scrapers which can be feed into Open Elastic Search Platform to built the search engine.
-
We have used various Open-Source and Cloud tools to streamline the data using Spark and Kubernetes to Elastic Search to create heavy loaded indexes for business.
-
Achieved tremendous output from the search engine which made the platform versatile to be used across various domains of business.
Mobility Data Management Studio
-
Data capturing and transformation of Mobilise devices are always difficult. We have come across the requirement to stream the data from mobilise object to take in consideration their consents and preferences.
-
Built a real time notification system which take consents & preferences from various mobilise devices & identify the notification popping for agreed objects.
-
We have developed an outstanding real time streaming pipeline which takes in the data through objects to open source streaming engine Kafka, make the choice based on the consents and preferences and delivers notification.
Machine Learning Ops
Automated scalable deployment of large recommendation engine to multiple markets
-
Large scale recommendation engine used by a major food chain client with global presence to benefit their retail franchise owners to recommend the best deal in their area/city.
-
Design and development of scalable production pipeline via code to organize, visualize and troubleshoot Kubernetes applications running on Amazon EKS across multiple market and regions.
-
Managed, organized and scheduled ML workflows & ETL jobs on clusters using Airflow. Applied best practices resulting in automate rapid and reliable delivery of new code changes with CI/CD pipeline.
Multimedia Critics Analytics with Social Influence
-
We got an urged from the customer to improve multimedia critics based on the location, team and features which can contribute to sentiments from social media.
-
Created an autonomous MLOps pipeline which provided end to end data delivery at an ease.
-
Capturing the data from various reviews and social media sentiments to a Machine Learning model which can predict the value of the media to its relevant users.
-
Managed the ML Engine, Data Pipelines and Data Quality Check to delivery with easy to use platform.