The client was planning to use a Linux based device for eLearning targeted at rural India. The Linux kernel had to be optimized, for fast boot time, operation speed and low power consumption. At ZeroOne, we removed all the unnecessary services and components and did a custom Linux build. The boot time was reduced by allowing only minimum services to run during boot time; other services were made to start in parallel in the background. During run-time, no additional services ran and those that were unused for a certain time would switch to sleep mode. This helped reduce the power consumption. For accessories, which did not have native support, an infrared remote for example, custom drivers were developed.

A multinational bank had numerous employees who travelled globally. The latter were having trouble accessing their data from the common server when they tried to login from other locations. ZeroOne was asked to understand the issue and solve this login problem. It was understood that due to overloaded Windows login scripts, the users potentially took almost 15 minutes to login to the system. The reason being - each country in the bank’s network used their own individual login scripts based on local policies. This lack of standardization had to be overcome.

The solution was arrived at in the following manner:

• The login scripts from different locations were collected and a pattern matching algorithm was run to identify the common scripts across the locations.
• These common and must have policies were programmed to load first, during a login

• Other must-have policies were made to load in the background, so that the user does not have to wait for the entire process to login and start work.

• This redesigned script gave the country managers the option to load any additional scripts/policies specific for that location, which would also load in the background.
• Thus, the boot time reduced from double digit minutes to single digit.
• The pattern matching algorithm was given as a web application to the client for future use.

Opening an account in a bank, the client, had multiple stages that were carried out by different people/departments. This caused a considerably large turn-around-time, which was not acceptable. There were manual tasks involved, such as data entry, physical file transfer, manager validations etc. All these processes had to be optimized to decrease the time taken to open a new account.

A web application was provided, by ZeroOne, to the data entry operators. The application had a tabbed interface, an Ajax based system for low bandwidth usage, thus enabling the operators complete the first stage quicker. The application also had an in-built feature, where complete verification of the details entered could be verified, using a maker-checker concept. The next steps in the process were directly transferred to the core banking system to carry out the final account creation process, overall reducing the account opening time significantly.

Within a bank there are many departments that handle a wide and varied range of tasks. In spite of the variation in the tasks, everything is interconnected and inter-dependant. The client bank, found that the communication between these departments, quite often, caused a bottle-neck in the completion of tasks. Data requests, Reminders, Escalations were constantly exchanged and very often missed out. ZeroOne was asked to step in to iron out these issues. Query management software was thus created and deployed across the bank branches to keep track of all the communication and requests that moved between departments.

The key to being a successful company is to make sure your customers are satisfied. The client, a multinational bank, was no different. ZeroOne created a customized solution that helped track and manage all tickets that were raised by customers. Complaints were fed in directly into the software and CC executives could view all details, manage the complaint and feed in the information regarding the status, all in one place.

The client, an HR recruitment firm, was having problems consolidating the resumes that they received across their branches. They found that the resumes, sent to their employee email ids did not always make it to the common pool. There was no central system that handled all the resumes.

ZeroOne analyzed the problem and brought together a solution as below:

• The patter matching algorithm was customized to identify the resumes files in a computer system.
• This pattern matching engine was run as a service on all the employees’ computers and it periodically searched for resumes in emails (Microsoft Exchange) and the hard disks.
• Files that were found were sent to a central server.
• The application had the intelligence to check for false positives and also identify duplicate resumes from multiple locations.
• The pattern matching algorithm was self learning which made the resume identification improve over time. It also learnt new resume formats.
• Since the data transfer was going to be huge, the architecture of the solution was based on intermediate store and transfer concept. Each state/ department can have intermediate nodes which will consolidate the resumes from the agents, compress them and send it to the main server.
• The time of searching, data transfer was configurable to the last level.
• Each document identified as resume was given ranks. If the rank falls in the doubt area, agents/ users can view the document and confirm if it’s a resume or not. The pattern matching engine auto-learnt from these decisions.

The client, a service company, wanted to keep track of their agents’ productivity as they found response time to a customer complaint was poor. Agents were found to not go on time, entering wrong information about the task or in some cases marking the query complete without visiting the customer location. Query closure by data entry was also insufficient.

To solve the problem, ZeroOne created a mobile app. Each agent was provided with a low cost android phone with this app loaded. The app will sync with the server on daily basis and show the tasks for each day. Each machine, at the customer’s place, was allotted a sticker with a QR code (to be replaced by NFC stickers soon). The agent has to scan the QR code using the mobile phone’s camera. The app will identify the machine and show the problem and the required fields to be entered. The app also has the feature of working offline after the day’s sync. Once the reading is taken, the app will send the details back to the server based on network access. The app will also record the time and location (based on GPS signals).This helps ensure that the agent has actually been to the location. Without the QR code there is no way of entering the data. QR stickers will be soon replaced by NFC stickers since the later is quite difficult to replicate. The app has the support for identifying each NFC sticker.