Freelance Senior Data Scientist

5/2017 – offen

● Built a Text analytics pipeline, leveraging STORM, kafka, hadoop and nltk for parsing
huge files and identifying named entities in them.
● Deployed Video analytics (voice and vision based NN models) for predicting (and
visualizing) emotions during the course of an interview.
● Helped the Pre-sales team, with writing proposals for RFPs, making them more
technically sound and sensible, along with helping them with the costing.
● Built a pre-sales demo with NN-based models for audio-cleaning tools with GAN
based NN model.
● Built a kidney stone size estimation tool for online assistance of kidney stone
lithotripsy procedures

Data Science, Maschinelles Lernen, Python

9/2005 – 3/2017

Senior Data Scientist – Greytip HR Solutions Aug.2015 – Mar.2017

● Deployed dashboards with the metrics for monitoring attrition/account loss and economic health metrics (Monthly Recurring Revenue, Customer Lifetime Value, Customer Acquistion Cost etc..).
● Built and Demonstrated a couple of predictive models to predict upsells and account loss probability.
● Designed, guided and helped setup end-to-end data warehouse (from distributed data sources), data pipeline, data-sanity checks, data dashboards and data modeling notebooks.
● Built a randomforest model to predict customer attrition and reduced attrition rates, by designing pro-active customer engagement process.
● Demonstrated a couple of predictive models to predict upsells and account loss probability as POC.
● Implemented a Reservoir sampling mechanism (with SQL triggers & random) and a generic SQL generator for the same.

Software Architect – Tagos Design Innovations Feb.2015 to Aug.2015

● Reduced annotation time by half, by automating parts of the video annotation task
● Reduced video segment loading time by the annotation tool, by replacing the
technology used.
● Created a web based tool to trim black bars(cinemascope) from images, using
imagemagick and tornado
● Built a QA tool for easier visualization and cross-checking (statistics of annotated(multi-dimensional) data, preview of touchable objects etc..)
● Use scrapy tool, to periodically scrape a page and automate extraction of new video as it arrives.

Senior Software Engineer – Chatimity Software Dec. 2013 – Oct.2014
● Scaled(from 1 to 2 Mn users) the backend to handle responsiveness demands
● Built a Retention analysis tool for checking and verifying weekly retention rates and helped decide features to build based on that
● Improved the product(chat-app) quality(latency and content quality) by improving backend services, analytic tools, and some probabilistic algorithms for auto-moderation of images based on user flags.
● Built a (tf-idf based for relevancy) ranking system for posts, based on recency, rarity and length(of text) in the post.
● Increased User engagement, by building a chat-bot to play word-games.
● Reduced new devbox setup time from a day to 1-2 hrs by automating setup and
dependency installation of a new system.
● Helped Devops monitoring by writing a parser for pushing sar command output into a graphite system.

Product Engineer – Commonfloor Techonologies Jul. 2013 – Sep.2013
● Ideated, designed and built a graph-based search engine for properties as POC.
● Flask based UI for downloading a image from a url, rotating it and uploading to S3 bucket.

Technology Lead – CSSCORP Technology Oct . 2011 – May.2013

● Reduced the time taken to switch router configs to one-tenth. Automated routing configuration changes in the network setup to load-balance(and re-route) during DDOS attacks or heavy loads for some specific domains,using a combination of django + fabric + triggers mechanism to issue the ssh commands.
● Eliminated manual effort for cleaning up log files, by providing a configurable tool to do the same.

R&D Engineer – Webnotes Technologies Aug . 2010 – May.2011

● Reduced down-time in case of a server problems by setting up the database with replication for robustness,
● Improved response time and reduced number of server-down incidents (even during peakload hours) by tuning mysql.
● Improved response latency with a module manager for importing and executing python modules from the filesystem, querying db for pyc(python bytecode) files.

Software Engineer @ Whitewall Networks Jan.2010 – Mar.2010
● Implemented a access control package for managing users, groups, permissions, Machine Images, Volumes, Disks and Cost limits using Django-ORM.
● Wrote a module to schedule, trigger and monitor remote python functions(as new processes) using multiprocessing and apscheduler packages

Part-time Programmer @CBCS, Allahabad University Aug.2007 – Jul.2009
● Programmer for Stimuli Presentation software for eye-tracker system
● Statistical programming(R) for some analysis of data output of the eye-trackingexperiments

System Analyst @ IBM Global Sep.2004 – Feb.2007
● Reduced the year-end balancing of fiscal batch time by customized the Finance
Transaction session to allow upload of single balance transactions, currency-wise.
● Sales order Upload: Sales order upload was a program that reads sales order headers and lines from an ASCII text file, sorts them according to the Sales order header and then creates/upgrades the sales order and lines table in Baan.
● Baan V Customization:Report Layout modification for Sales Invoice layout of
Hungary,china,chile. As a consequence of the RoHS Guidelines ECR,there were a few fields that needed to be printed based on whether the RoHS flag was true or false.

Data Science, mySQL, Extreme Programming, Software Architektur / Modellierung, Python