Post Job Free
Sign in

Data Software Engineer

Location:
Gurugram, HR, India
Posted:
April 04, 2017

Contact this candidate

Resume:

Akash Verma

Technology Analyst – Machine Learning

Gurgaon, IN Email: *****.*********@*****.*** Phone: +91-852******* Skype: akash.ql2

Innovator, Evangelist and Technical Entrepreneur

Experience in demonstrating the ideas with rapid scalable prototypes and selling them to business and engineering teams to move the prototypes to production.

Strong Machine Learning and Data mining Experience.

Deep analytics and understanding of Big Data and algorithms using Hadoop, Map reduce, NoSQL and distributed computing tools.

Architected and implemented several recommendation and personalization algorithms.

Implemented algorithms in Python.

Excel at extracting intelligent patterns out of huge amounts of structured and un-structured data.

Strong Start up experience

Strong Information Retrieval and web crawling/web scraping Experience. Experience in managing and leading multiple large scale projects.

Research Interests Spanning

Web Mining, Machine Learning, Information Extraction, Pattern Recognition, Artificial Intelligence, Genetic Algorithms, Neural Networks, Support Vector Machines, Fuzzy Logic, Bayesian Networks, Natural Language Processing, and Data mining

Software Technologies/Tools/Environment.

Python, Scikit-Learn, Regular Expression, MongoDB, SQL Server, Hadoop, NoSQL, Mapreduce, Pig, Hive, C++, Java, HTML, XML, MySQL, Eclipse, SVN, JIRA, Git,, Stanford NER, NLP

Willing to relocate: Anywhere

WORK EXPERIENCE

Technology Analyst – Machine Learning

Annik Technology Services - Gurgaon, IN - May 2016 to Present

Architect and implement machine learning/Web scraping algorithms.

Title: “Comprehensive Crawler based Named Entity recognition using Hidden Markov Model & Look Up”

Description: Developing a Named Entity Recognizer using Hidden Markov Model

Understanding making of several other fundamental decisions as what model to use for sequential inference, how to represent text chunks and what inference algorithm to use. Furthermore, this project automates the process of lookup and identification of the entities by linking to a comprehensive crawler with features like Restricting followed links, URL Normalization, Path Ascending Crawling and

Academic Focused Crawling. Thus this project lays the groundwork of developing a

Commercial product intended to develop a solution for several data and information mining setups.

Title: Task Tracker Prediction for employees

Description: Comment Validation for an internal task tracker tool.

Client: Microsoft Corporation

Title: “Web Search Automation for Microsoft’s Bing search Engine for Subsidiaries Search Bot”

Description: Search keywords on Bing search engine and extract the subsidiaries names of any organization using NLTK and Stanford NER.

Client: Microsoft Corporation

Title: Agreement Search Tool

Description: Create strings to search organization names of spelling mistakes or similar names and collect the related records by using partial ratio “gestalt pattern matching” algorithm.

Client: Canon Research Development Center

Helped prototyping and interfacing several startup APIs in developing an enterprise software product to extract intelligence out of huge amounts of digital data in the enterprise. Provided technical expertise on the feasibility/scalability of startup technologies. Attended big data related meetups and produced reports for the management.

Client: NPD

Title: New Launched Products Found from Websites

Description: We scrape the many websites like e-commerce and brand websites on daily bases and provide the information of new products found on website.

Sr. Software Engineer – Machine Learning

Magic FoodBox India Pvt Ltd – New Delhi, IN - Oct 2015 to May 2016

Title: The Frying Pan (A personalized food App)

Description: Information related to Recipes is extracted from various leading websites and transform them as required. Write machine learning algorithms for cuisines, meal category, recipe tags and serving size etc. Insert data to MongoDB through python/scrappy and also to S3 bucket through Django API with python script.

.

Application Developer

QL2 Software - Gurgaon, IN - March 2014 to July 2015

Client: Fortune 500

Title: Web Crawling, Machine Learning and Pricing Integration

Domains: Retail, Automobile and Tires, Petrol and Public storage, Telecom Billing Management

Description: All the relevant information that is available on the websites is extracted. Major clients are taking data from more than 200 websites. There are various leading online shopping websites from where data is extracted. Information extracted from websites according to the input provided. Input may be (keyword/Path). There are two types of applications available in Retail domain (I) that scraps data from the list pages and (II) that scraps data from detail pages. Here data being extracted and further processed according to Client’s requirements for pricing integration and then delivered.

Client: Fortune 500

Title: Sentiment Analysis on Product Reviews

Domains: Retail

Description: We extract product review comments from client website as our data set and classify text by subjectivity/objectivity and negative/positive attitude. We propose different approaches in extracting text features such as bag-of-words model using large product reviews corpus restricting to adjectives and adverbs, handling negations, bounding word frequencies by a threshold, and using WordNet synonyms knowledge. We evaluate their effect on accuracy of four machine learning methods-Naive Bayes, Decision Trees, Maximum-Entropy, and K-Means clustering. Then, We provide the review information to client like review id, review text, product id and review analysis like positive or negative.

EDUCATION

Bachelor of Technology in Computer Science & Engineering

UPTU - Lucknow, IN

July 2013



Contact this candidate