Abhishek Karanth

Machine Learning Engineer

Bengaluru, Karnataka, India7 yrs 8 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Expert in Machine Learning and Data Visualization.
Successfully improved forecasting accuracy in airline industry.
Proficient in big data applications using Scala and Spark.

Stackforce AI infers this person is a Data Scientist specializing in Machine Learning and Big Data applications.

Contact

Skills

Core Skills

Machine LearningData VisualizationBig Data DevelopmentTime Series ForecastingSoftware Development

Other Skills

PythonSQLNeural NetworksScalaSparkSeabornPlotlyData ProcessingETLR-ShinyRuby on RailsTest-Driven DevelopmentRubyGitPostgreSQL

About

● Data Scientist with 6+ Years of total IT work experience in Amadeus Software Labs. ● A blend of Data Science and Software Engineering, capable of end to end development of Machine Learning models. ● Over 4+ years of work experience as Data Scientist having expertise in Machine Learning, Time Series Forecasting, predictive modeling, data analysis, Statistical modeling, data mining, big data development. ● Extensive hands-on experience of using Python Programming including numpy, pandas, scikit-learn, scipy, statsmodels, matplotlib, seaborn modules along with exposure to Tensorflow module for solving business problems using advanced Machine learning and Forecasting techniques. ● Hands on Expertise in Software Development with expertise in Algorithms, Data Structures, OOPs, Python Programing. ● Hands on Expertise in Supervised Problem (Regression and Classification) such as Linear/Logistic Regression, KNN, Decision tree (CARTs), Random Forest, Adaboost, Gradient Boosting, XGBoost, Light GBM and Unsupervised Problem (Clustering) such as KMeans, Hierarchical Clustering. ● Hands on Expertise in Time Series Forecasting Models such as SARIMAX, Exponential Smoothing and working knowledge on frameworks like Facebook Prophet. ● Analyzed and processed complex data sets using SQL queries and created reports and dashboards using Data Visualization tools like seaborn, plotly. ● Working knowledge in Apache Spark and building big data applications using Spark via Scala APIs. ● Productionized ML models by following best programing practices including Test Driven Development.

Experience

7 yrs 8 mos

Total Experience

3 yrs 10 mos

Average Tenure

4 yrs 5 mos

Current Experience

Meesho

Machine Learning Engineer 3

Nov 2021 – Present · 4 yrs 5 mos · Bengaluru, Karnataka, India

Machine LearningData VisualizationPythonSQL

Amadeus labs

3 roles

Member Of Technical Staff - Data Scientist

Apr 2020 – Nov 2021 · 1 yr 7 mos · Bengaluru, Karnataka, India

Adapted the existing models to mitigate the impact of COVID-19 on the airline industry and improved the MAPE from 35% to 17%.
Worked closely with the product team to build an airline fare prediction model using Neural Network with an overall MAPE value of 15%.
Developed a big data application using Scala-Spark to parse around 40k files daily and extract information about flight delay in 15 minutes. Also added the capability of historical reprocessing and optimized the application to process entire year in an hour. Entire work from Design to Code was managed by me.
Developed basic interactive dashboards and charts using Seaborn, Plotly to explain the trends and recovery in airline industry from the impact of COVID-19.
Conducted and lead the Amadeus wide data science contest in 2021, starting from the formulation of the problem, evaluation metrics till the final jury round.
POC on Passenger Segmentation using Principal Component Analysis (PCA) and K-Means Clustering.

Machine LearningNeural NetworksScalaSparkData VisualizationSeaborn+1

Senior Software Engineer - Data Scientist

Promoted

Apr 2017 – Mar 2020 · 2 yrs 11 mos · Bengaluru, Karnataka, India

Implemented and productionized a generic passenger forecasting model using Time Series Forecasting Techniques such as SARIMAX, Exponential Smoothing for 25k different time series with overall MAPE of 10%. Used parallel processing to reduce the processing time by 15x.
Involved in building a model based on the Gradient Descent algorithm to come up with true airline passenger count across the world with a MAPE of 6% at leg/leg-airline level.
Implemented and productionized a mathematical model from scratch to come up with connecting passengers share at an airport using Curve Fitting.
Developed many ETL and Data Processing tasks using pandas and numpy.
Mentored an intern and provided guidance in building Quality Check Dashboards.
Built an interactive validation dashboard using R-Shiny with quick response time even for a large dataset.
Owned the Python Web Services after getting handover of the product and implemented new features using Flask.

Time Series ForecastingPythonData ProcessingETL

Software Engineer

Jul 2015 – Mar 2017 · 1 yr 8 mos · Bengaluru, Karnataka, India

Worked on Ruby on Rails Web Services as a single developer from the local site with best software practices such as Test-Driven Development.
Developed a common Web Services component to achieve scalability of the product.
POC for new Scalatra Web Services technology and involved in the implementation of the foundation of the same.

Ruby on RailsTest-Driven DevelopmentSoftware Development