Arindam Das
Data AnalystAI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI Architect & Valeo Expert in AI at Valeo | PhD Student at University of Limerick | Computer Vision & Deep Learning | Autonomous Driving
▪ With 2 years of experience as an Artificial Intelligence Architect, successfully leading and guiding multiple teams across various sites, delivering ML-based products in several geographical locations. ▪ With 7.5 years of dedicated experience in automated driving systems, consistently focused on delivering solutions for major automotive OEMs. ▪ 11 years of comprehensive experience in computer vision and deep learning algorithms, developed solutions for various real-time products. ▪ Expert in research, developing AI algorithms and models that are suitable for edge devices. ▪ 52+ patents filed in the field of AI for automated driving, printing, document analysis. ▪ Google scholar: https://scholar.google.co.in/citations?hl=en&user=W8DTl_gAAAAJ For any query or clarification, you can reach me through mr.arindam.das@gmail.com
West Bengal University of Technology, Kolkata
Bachelor of Technology (BTech), Computer Engineering
January 1, 2009 – January 1, 2013
Uttarpara Amarendra Vidyapith
High Schooling, Mathematics, Physics Chemistry and Biology
January 1, 2007 – January 1, 2009
Valeo
AI Architect
April 1, 2023 – Present
Valeo
R&D Lead
January 1, 2023 – April 1, 2023
Valeo
Senior Lead Engineer
August 1, 2022 – December 1, 2022
Valeo
Expert in AI
January 1, 2022 – Present
Valeo
Lead Engineer
April 1, 2021 – July 1, 2022
Valeo
Senior Engineer - Deep Learning
January 1, 2018 – March 1, 2021
HCL Technologies
Lead Engineer
July 1, 2016 – January 1, 2018
Greater Chennai Area
HCL Technologies
Software Engineer
February 1, 2015 – June 1, 2016
Greater Chennai Area
Cognizant Technology Solutions
Programmer Analyst
February 1, 2015 – February 1, 2015
Cognizant Technology Solutions
Programmer Analyst Trainee
February 1, 2014 – February 1, 2015
Indian Statistical Instiute, Kolkata
Research Assistant
October 1, 2013 – February 1, 2014
Greater Kolkata Area
Indian Statistical Instiute, Kolkata
Intern
July 1, 2012 – October 1, 2013
Greater Kolkata Area
Indian Statistical Instiute, Kolkata
Natural Language Processing
June 1, 2011 – October 1, 2011
Greater Kolkata Area
Generalized Stacking and Intra-Domain Transfer Learning Based Document Image Classification Using DCNNs
August 1, 2017 – January 1, 2018
In this work, a region-based Deep Convolutional Neural Network framework is proposed for document structure learning. The contribution of this work involves efficient training of region based classifiers and effective ensembling for document image classification. A primary level of ‘inter-domain’ transfer learning is used by exporting weights from a pre-trained VGG16 architecture on the ImageNet dataset to train a document classifier on whole document images. Exploiting the nature of region based influence modelling, a secondary level of ‘intra-domain’ transfer learning is used for rapid training of deep learning models for image segments. Finally, stacked generalization based ensembling is utilized for combining the predictions of the base deep neural network models. The proposed method achieves state-of-the-art accuracy of 92.2% on the popular RVL-CDIP document image dataset, exceeding benchmarks set by existing algorithms.
Various Illuminance Affected Font and Color Independent Text Binarization of Camera Captured Images
November 1, 2015 – May 1, 2016
This article presents our recent study on multi colored text binarization. In the output image, we represented foreground content as black and background as white regardless the polarity of foreground and background in original image. Here we applied connected component analysis based approach to group the words or characters within bounding or edge box. The main novelty of this reported work includes the calculation of each edge box based local color threshold value from CIELAB color space. This approach makes the proposed system capable of binarizing multi colored texts where a single character has more than one color. The proposed method has been executed on well-known D1BCO2009 and CMATERdb datasets that contain a large set of images to show the efficiency over other existing methods through qualitative comparison study.
Automatic Number Plate Recognition With OCR
March 1, 2015 – June 1, 2015
An intelligent system has been proposed for vehicle number plate recognition system with complete OCR. Here some challenges have been met with excellent output like multiple number plate recognition, successfully dealing with brightness problem on number plate etc. After number plate localization and extraction an OCR system has been integrated in order to get the alphanumeric content of the number plate in file. For recognition purpose Support Vector Machine has been employed with features which showed 95.72% performance accuracy. Other than OCR result, the performance of number plate extraction and character segmentation has been found to be 97.1% and 95.4% respectively. The perfomance evaluation of the proposed system has been carried out on our own dataset of 560 sample images.
Offline Handwritten Character Recognition in Bengali Script
June 1, 2013 – February 1, 2014
Currently I'm working on it under the direct supervision of Dr. Ujjwal Bhattacharya at Handwriting Recognition Lab of Computer Vision and Pattern Recognition Unit at Indian Statistical Institute, Kolkata. To develop a robust recognition system for handwritten characters need an extreme and exceptionally good pre-processing steps and among those steps, slant estimation and correction is one of them. My main focus area was to estimate and remove slant from handwritten characters. After successfully coming up with word level slant removal I approached for whole document level slant removal without processing the document in each word level. As far as our knowledge goes from literature review, this is the first approach to estimate and remove the slant without word level segmentation. Here whole document can be processed without any information loss. Primarily I was working on Bengali script but later this system has been tested on good number samples of Devanagari, English, Russian and Georgian scripts and showed impressive performance. This system has also been tested on ancient handwritten documents. The publication process of this proposed method has been undertaken.
Skew Detection and Correction in Bengali Script
January 1, 2013 – July 1, 2013
Skew detection and correction is another challenging task in achieving successful character segmentation and henceforth recognition. Here skew detection and correction has been implemented using Hough Line transformation. Skew up to 60 degree has been detected and successfully corrected.
ISIgraphy
September 1, 2012 – January 1, 2013
ISIgraphy, the first Android application developed at the Handwriting Recognition Lab (www.isical.ac.in/~ujjwal) of the Computer Vision and Pattern Recognition (CVPR) Unit of Indian Statistical Institute, Kolkata. This application can be used to collect and store handwriting samples. The sample is stored in the user given filename. The stored handwriting data can be later retrieved and displayed to view or send it by email to someone. This software will be particularly useful to the researchers/practitioners working on handwriting analysis. Such people may use the software to collect handwriting samples on their Android-based devices. This software supports multiple users on the same device.
SUDOKU - A Game of Fun & Intelligence
June 1, 2011 – Present
• It was a project based on conventional gaming method of 9x9 SUDOKU using Graphics and implemented in C. • The game has three stages those are easy, medium and hard. After solving one Sudoku puzzle the probability of occurring same puzzle is nearly zero. Though it is an estimated and experimented result.
Cultural Fit Analysis
The candidate's background is heavily skewed towards AI, Computer Vision, and Deep Learning, with significant experience in R&D within the automotive and printing domains. While these skills are valuable, the target role is 'Data Analyst'. The candidate's experience is more aligned with a Data Scientist, Machine Learning Engineer, or AI/Computer Vision Engineer role. The project diversity is strong within the AI/CV domain, but there's less explicit evidence of traditional data analysis, business intelligence, or broader data science projects that might be expected for a typical Data Analyst role. This suggests a potential mismatch in the specific focus of the target role versus the candidate's deep specialization.
Soft Skills & Operational Fit
The candidate's project descriptions highlight a strong research orientation, problem-solving capabilities, and a drive for innovation, as seen in their work on novel slant removal techniques and document image classification. Their progression through various lead and expert roles at Valeo suggests leadership potential and the ability to drive technical initiatives. The detailed project descriptions indicate good technical communication skills, though no specific soft skill assessments were provided.