Create a domain taxonomy from the unseen documents.

My Role:
➢ Reading the papers and work with architecture to implement the auto
taxonomy creation from the unseen documents.

Rapid Red Flag(RRF) checks quickly helping you to select the
most suitable candidate for your business by checking the Risk score for the
following checks based on the Natural Language Processing techniques,
1. Litigation Check
2. Adverse Media Check
3. Sanctions Check
4. Background Check
My Role:
➢ Work on the analytics part of the modules which includes Taxonomy
creation, generating WSD corpus, sentiment analysis, Risk type
classification, clustering etc.
➢ Understand and implement the requirements with Architecture
support.
➢ Handling the client requirements.
➢ Worked on UI.

Information extraction from the PDF/word documents. By
using the DEFT Framework, extract the following features
➢ Named Entities(NER)
➢ Entity Relations(SVO’s)
➢ MultiWord Expression
➢ Semantic Extraction
➢ Semantic Roles(SRL) extraction
➢ Key Terms and their rank Extraction
➢ Coreference
Create a Rest End Service for the above features for the consumers. By using
OrientDB we can represent the output in graphical view and to use it as search
query from the Orientdb.
For the scalability, Same framework implemented in Hadoop in AWS cluster
environment by using Spark.
My Role:
➢ Work on most of the features listed above
➢ Handling the scrum meetings and conduct Retro meetings.
➢ Guiding the new Engineers.
➢ Support in running DEFT in Hadoop environment

Annotate the given customer data and generate the training
data. Generate the custom NER model with the new entity types given by the
customer. Identify those entities from the extracted PDF/word documents and
replace with the entity types. Generate the cross-validation matrix to get the
accuracy of the model.

BAT is tobacco company. Analyze the data and build a
classification model for auto categorization and work assignment through
various models and predict the tickets are raised getting classified correctly.

Platform software of Boeing 737 Max is combination of
Bootloader, Device Drivers, Core BSP, Flash File System, Data load,
HMA(Health Monitor) and TFTP library.
My Role:
➢ Analyze, design, develop and unit test using C for TFTP library.
➢ Involved in all phases of Software Development life Cycle for this
module.
➢ Conducting team meetings and monitoring progress of work.
➢ Mentoring New Engineers and giving knowledge share sessions.

HUD is any transparent display that presents data without
requiring users to look away from their usual viewpoints. The origin of the
name stems from a pilot being able to view information with the head
positioned "up" and looking forward, instead of angled down looking at lower
instruments. Typical aircraft HUDs display airspeed, altitude, a horizon line,
turn/bank a slip/skid indicators.
My Role:
➢ Analyze, design, develop and unit test using C++.
➢ Involved in all phases of Software Development life Cycle for this
module.
➢ Conducting team meetings and monitoring progress of work.
➢ Mentoring New Engineers and giving knowledge share sessions.a

Onboard Maintenance System is an application which will
monitor the faults and generates the reports upon user request. Mechanic will
use these reports as a reference to fix the root cause of the failures when the
aircraft is landing on the ground and Pilot can also use these reports incase
when he needs to know the detail information about the faults.
My Role:
➢ Implemented the code in C and part of the Integration team to test the
applications on target environment.
➢ Involvement in NCT, VCT, PCT.
➢ Familiar with writing software Remote procedure call in distributed
process communications (using XML-RPC libraries)

In SVN, each repository contains branches, tags, trunk. Project contains so many applications and libraries and each contains so many branches and tags. when project checkout it will checkout all branches, tags and trunk but all are not necessary to check out for a programmer. It also takes a lot of time to checkout (based on project size) and occupies lot of system memory. To save checkout time and system memory, I come up with this lean idea.