MS in Data Analytics (Dublin Business School) Ireland.Having strong concepts and intermediate-level knowledge on Statistics,Machine Learning,Data Mining, Data Visualization, Data Warehousing,NoSQL, R,Python, Deep Learning.Worked on many academic projects.
DevOps Engineer with 7 years of experience in Investment Banking Industry. Good experience in trading tools OTC as well as exchange based(Buy-Sale Both Side) ,Risk Management Tools,DevOps tools like Perforce, Git-bucket. (Trading Support Specialist with FIX protocol).
-
Experience
PROFESSIONAL EXPERIENCE
ASSOCIATE -- DEUTSCHE BANK (2016-Jul --2019-Dec)
Foreign Exchange (Global Markets)-High Frequency/Ultra Low Latency Algo Trading
Real time Market Data Analysis with incident/change/problem management - responsible for Market Data/Pricing distribution to various trading applications used across DB.Market Making,Market Connectivity,High Frequency Algorithmic Trading,Client FIX connectivity,Pricing to huge no. of clients,Auto Hedging.It utilizes facilitating ECN and Exchanges connectivity through complexcircuits/links,market data publisher to various trading applications used across DB (FX,listed derivatives,rates and credits,etc) .Providing efficient technical support service to Traders/TA’s,Desks/Quants,BU,clients (Buy amp; Sell Side both) RAPID is a low latency high frequency spot trading platform which is responsible for almost 80% of the FX trading across DB.RAPID provides functionalities like Algotrading, Autohedging, pricing,spread calculator/multiplier,Risk engine,Position keeping,Execution Manager,Strategies/Algorithms Management,Order Manager, etc. Managing external clients FIX connectivity through FIX gateways( FIX protocol 4.4 and FIXT1.1).
Awards/Recognition : “Spot Excellence” Award 2 times
APPLICATIONS ENGINEER --SUNGARD FINANCIAL SERVICES (Fidelity Investment Solutions)
2015-Sep -- 2016-Jul Sungard Global Trading - Standard Chartered Bank, JP Morgan and OCBC. (SLA Based)
SUNGARD Global Trading team is responsible for providing Real time technical support with incident/change/problem management to Trading GL applications support to clients Including Incident,problems and change management. Mainly focusing in solving the business users queries related with the trading which includes buy/sell orders, market making also provides service desk traders taking orders from clients to place order to global markets on their behalf.Throughout the day there are various files that need to be exchanged between Sungard and the clients.
SYSTEM ENGINEER -- TATA CONSULTANCY SERVICES (2012-Dec -- 2015-Sep)
Sales and Trading Fixed Income Derivatives/Cash (Global Makets) -Morgan Stanley
Real time production support(technical) with incident/change/problem management services to Traders/TA’s,Desks,BU&Clients.DevOps Team is responsible for exchange and OTC based trading activities of the derivatives like option,bonds,swaps,futures,etc.Reconciliation of the trades from upstream to downstream (various risk system).Real Time Risk and PnL calculation performed by different engines for derivatives.Marking and cooking of the curves including Desk and Traders EOD sign off for the EOD calculations.Involvement of the Strategists for calibrations of Interest rate and yield curves, reporting calibration issues with the knot point of the curves.Perforce-Version controlling tool for doing slurm related changes,code enhancement for zapp.
Awards/Recognition: Received “Performer of the Year” award.
-
Projects
ACADEMIC PROJECTS:-
-Visualize natural groupings or clusters in (sign_mnist and customers dateset) using dimensionality- reduction & clustering algorithms(kNN), TSNE/UMAP/PCA - Python
PCA Vs t-SNE Vs UMAP
PCA – By taking look at the plot, we can say that it didn’t work well on segregating the clusters. Even though it generates the output very quickly, it is not feasible to go ahead with this approach. Because it is generating very dense clusters from which identification of clusters will be tedious task. However, it will be very effective when it comes to the smaller datasets clustering.
t-SNE – It has managed to do good work on segregating the clusters, since the visualization is better than PCA. However, it is computationally very expensive as it takes too much time to produce the output.
UMAP- By looking at the graph, we can easily conclude that UMAP is doing better in separating the signs as compared to PCA and TSNE.UMAP turned out to be the most effective manifold learning technique in terms of displaying the different clusters. As some of the outputs that we have observed during the runs were well defined in terms of clustering. And it is significantly faster than t-SNE technique.
-How Netflix Uses its per title encoding algorithm to create the representation best suited to consumer bandwidth - Python/Linux
FFmpeg/python/linux
Our aim is to draw a conceptual graph, which illustrates the
relationship between the quality and bitrate for a sample video
source encoded at different resolutions.
Three resolutions i.e. 180p, 360p and 720p are used to choose the
resolution, which is best, suited for representation. Firstly, RD
curve for each resolution has been drawn .
In order to perform mathematical analysis on the optimal
representation, polynomial fitting is used.The Bitrate-PSNR pairs of respective resolutions calculated .
-Dynamic Retail Dashboard for local retail marts in Ireland - Tablue
-Detect Credit Card Payment Default By Customer Using CRSIP-DM Methodology - RapidMiner
The motive of this project is to predict,weather the Credit Card Customers will default in making payment next month or not. With idea of how the financial service providers, keeps the track of fraudulent credit card transactions, so that they can avoid unnecessary charges to the customers for the items they didn’t purchase. Insights from predicting the model can help financial institutions in forecasting the change in market demand and decision making; at what time a new product can be launched.
-Proposed dissertation project -Video Salience using openCV(python)
For your ref -->https://github.com/wakdamo
CERTIFICATION:-
ITIL Certified
Machine Learning –Deltiin India Tech Pvt Ltd.
Financial Markets - Bombay Stock Exchange Ltd.
Interest Rate Deriva