Saikat Goswami, Developer in Bengaluru, Karnataka, India
Saikat is available for hire
Hire Saikat

Saikat Goswami

Verified Expert  in Engineering

Machine Learning Developer

Location
Bengaluru, Karnataka, India
Toptal Member Since
August 22, 2019

Saikat has several years of experience as a software engineer in the credit card and payment domain. As a self-taught ML engineer, he's passionate about solving complex problems using machine learning (ML). Saikat communicates exceptionally well and has led several projects in his career.

Portfolio

Freelance Clients
Pandas, Python, Git, Deep Learning, Data Science, APIs, Microsoft Excel...
VISA
REXX, IBM z/Transaction Processing Facility (z/TPF), C#, NumPy, Pandas, SciPy...
RS SOFTWARE
REXX, IBM z/Transaction Processing Facility (z/TPF), Java, C#

Experience

Availability

Part-time

Preferred Environment

Pop!_OS, Visual Studio Code (VS Code), Git, Jupyter Notebook

The most amazing...

...ML project I've worked on is a Spotify music segment recommender system using K-means to cluster similar music and nearest neighbors for recommendations.

Work Experience

Software Developer

2018 - PRESENT
Freelance Clients
  • Developed a deep learning model on object detection using the TensorFlow Object Detection API and SSD Mobilenet model.
  • Built a Spotify music-segment recommender system using the k-means clustering algorithm.
  • Created a deep learning classification model on a DICOM dataset.
  • Deployed several web scrapers using BeautifulSoup, Selenium, and Scrapy.
  • Developed a Python framework to support building intraday trading algorithms on a platform.
  • Developed a program to continuously scrape data from Betfair and Odds Portal websites and aggregate both data sources.
Technologies: Pandas, Python, Git, Deep Learning, Data Science, APIs, Microsoft Excel, Machine Learning

Senior Software Engineer

2015 - 2018
VISA
  • Ideated and implemented a machine learning model for the risk identification of regular installation loads.
  • Implemented a machine learning model for triggering various regression runs depending on the defects identified.
  • Developed the implementation of a tokenization infrastructure.
  • Spearheaded several projects on tokenization and various automation products.
  • Designed and implemented an end-to-end solution to compare transaction logs between codebases.
Technologies: REXX, IBM z/Transaction Processing Facility (z/TPF), C#, NumPy, Pandas, SciPy, Scikit-learn, Machine Learning, Python

Software Engineer

2013 - 2015
RS SOFTWARE
  • Created a testing framework for automated transaction execution and report generation.
  • Built an automated tool for the batch validation of authorization transaction results which reduced the execution time.
  • Constructed a tool for generating tokens on the fly for provisioning.
  • Developed a tool to read and modify global tables in TPF to accommodate token-transaction processing.
  • Implemented an end-to-end solution to read data from logs and extracts transactions from test buckets.
Technologies: REXX, IBM z/Transaction Processing Facility (z/TPF), Java, C#

Associate Software Engineer

2011 - 2013
RS SOFTWARE
  • Participated on a multiregional team on the Oracle ERP system for the P2P cycle.
  • Generated regular and ad-hoc reports. Troubleshot peripheral tools.
  • Analyzed and developed code patches.
  • Automated the system build process which brought down the required time from 30 minutes to no more than three minutes.
Technologies: Toad, Microsoft SQL Server, Java, PL/SQL, Oracle Apps

Spotify Music Segment Recommender

https://github.com/sgsaikat/Spotify_Music_Recommender
This project uses an artist's name as input and recommends song clips of similar artists based on musical similarity for novel music production using a K-means clustering algorithm.

Speech Recognition

https://github.com/sgsaikat/Speech_Recognition
This project deals with a multi-class classification problem which takes audio data as input and identifies the text/word using mel-scaled spectrograms and a CNN-based deep learning model.

Tweet Sentiment Analysis

https://github.com/sgsaikat/Sentiment_Analysis_Tweets
A binary classification problem to identify positive/negative sentiments from Tweet data using Word2Vec and an LSTM model.

Time Series Forecasting

https://github.com/sgsaikat/Beijing_PM2.5_Time_Series_Forecasting
This deals with continuous weather data forecasting using a Beijing PM 2.5 dataset along with an LSTM model.

Music Generator

https://github.com/sgsaikat/Music_Generation
This project generates ABC music notations using an LSTM model.

Face Recognition

https://github.com/sgsaikat/Computer_Vision/blob/master/Face_Recognition.ipynb
This project creates a database of known faces and labels them. A person can identify unknown people in photographs and videos by using this database.

GeoCoder App

https://github.com/sgsaikat/GeoCoder
This web application extracts addresses from the uploaded file and determines the geocodes for the same with an option to visualize the data on a map.

Lunar Rock Classifier

https://github.com/sgsaikat/Lunar_Rock_Classifier
This is the code for Lunar Rock Classification problem in HackerEarth.

Web Scraping Amazon.in

https://github.com/sgsaikat/WebScraping_Amazon.in
This web scrapes the Short Reads section of Amazon.in.
2007 - 2011

Bachelor's Degree in Electronics and Communication Engineering

JIS College of Engineering - West Bengal, India

OCTOBER 2022 - OCTOBER 2025

AWS Certified Developer - Associate

AWS

DECEMBER 2019 - PRESENT

The Python Mega Course: Build 10 Real World Applications

Udemy

AUGUST 2019 - PRESENT

Applied ML Course

Applied AI Course

MARCH 2019 - PRESENT

Deep Learning and Computer Vision A-Z

Udemy

MARCH 2019 - PRESENT

Python for Data Science and Machine Learning Bootcamp

Udemy

JUNE 2016 - PRESENT

Big Data Hadoop and Spark Developer

Simplilearn

SEPTEMBER 2015 - PRESENT

Programming for Everybody (Getting Started with Python)

Coursera

Libraries/APIs

Scikit-learn, Pandas, NumPy, Keras, PySpark, OpenCV, SciPy

Tools

Git, Toad, AWS Glue, Microsoft Excel

Paradigms

Data Science

Frameworks

Scrapy, Flask, Serverless Framework

Languages

Python, Python 3, SQL, REXX, Java, C#

Storage

Microsoft SQL Server, Databases, Oracle 11i, PL/SQL, Amazon S3 (AWS S3)

Platforms

Windows, Jupyter Notebook, Amazon Web Services (AWS), Visual Studio Code (VS Code)

Other

Data Cleaning, Data Handling, Machine Learning, Deep Learning, Software Development, Software Engineering, IBM z/Transaction Processing Facility (z/TPF), Pop!_OS, K-means Clustering, APIs, Web Scraping, Scraping, Data Engineering, Oracle Apps, Shell Scripting, Statistics, Mathematics, Linear Regression

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring