We build products to make us more efficient and to be more productive. Most of these products were only developed once we felt it would take too long to develop traditional applications.
For example, the CSV Loader was built to get data into a database without spending weeks building individual ETL applications per file. It has saved us weeks of time whilst allowing us to get on with more important tasks.
Right now, none of these products are for sale. We are using them on our own solutions, which is helping us to iron out flaws and inconsistencies during testing before we release them commercially.
We have been sourcing data from many different providers. Even providers have different CSV formats, we wrote an application that; Accepts a delimiter type > Accepts a folder source > Accepts a SQL Database Target > Imports the files into the database.
XML is a complicated data format to deal with. Many utilities exist for importing XML data, as long as it has been well-defined and well-structured. We wrote this application to get data from one or more XML files containing one or more tables into a database.
With extensive experience of developing on and working with Batch Management Frameworks. Systems such as Control-M, Autosys and Dollar Universe are well known in large financial institutions. Excellent at scale, but clunky and limited to seasoned developers.
To overcome this, we built a very lightweight .Net task discovery job runner framework.
This was a simple requirement - can we, at certain points, ask an application to zip up one or more folders, saving them to a location where they are incrementally saved? Again, we just wanted a simple task fulfilled.
It was a neat outcome and automatically manages file space use.
We will definitely want to have a better name than this. We found us sometimes; needing to extract data from a file, move a file based upon its content, delete a file for any number of reasons. Create a new file with a subset of content from a main file.
This application has become a labour of love for us. It will never be perfect, and yet, we can scan thousands of web addresses, pulling specific information to files for further analysis. We are not sure whether we will ever release it because of its imperfections and yet it works so well, we may just do.
NLP is computationally interpreting sentences and words to garner useful information, capable of inferring true meaning or, statistically categorising information. We are working on an application to let us extract meaningful information from search results.