Navaneethan Santhanam

Chief Scientist at Smartterra
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
Bengaluru, Karnataka, India, IN
Languages
  • English Native or bilingual proficiency
  • Tamil Native or bilingual proficiency
  • Hindi Professional working proficiency

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

5.0

/5.0
/ Based on 1 ratings
  • (1)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

Will Kurt

Navaneethan has the rare combination of creativity, technical understanding and strong communication skills that make for a great Data Scientist. At KISSmetrics he was known for having a great relationship with every team he worked with and was always able to come up with innovative solutions to data problems. Navaneethan stands out among Data Scientists and it was a pleasure to work with him.

You need to have a working account to view this content.
You need to have a working account to view this content.

Experience

    • India
    • Software Development
    • 1 - 100 Employee
    • Chief Scientist
      • Sep 2020 - Present

      Building AI for water utilities Building AI for water utilities

    • IT Services and IT Consulting
    • 1 - 100 Employee
    • Lead Machine Learning Engineer
      • Oct 2019 - Sep 2020

    • Senior Machine Learning Engineer
      • Dec 2016 - Oct 2019

      I build machine learning systems for natural language processing from noisy channels such as social media and chat. I have worked on several production-ready systems that help brands optimise their customer support systems and make them more efficient and reliable Identifying ‘actionable’ tweets and Facebook posts from brand handles and pages • Built a word2vec-based model for predicting which posts/tweets need follow-up • Achieved ~90% accuracy on Facebook and Twitter… Show more I build machine learning systems for natural language processing from noisy channels such as social media and chat. I have worked on several production-ready systems that help brands optimise their customer support systems and make them more efficient and reliable Identifying ‘actionable’ tweets and Facebook posts from brand handles and pages • Built a word2vec-based model for predicting which posts/tweets need follow-up • Achieved ~90% accuracy on Facebook and Twitter data Named Entity Recognition for Chatbots • Built a Named Entity Recognition (NER) system for identifying various named entities in chat • Achieved 90+% precision and recall in production • Exceeded customer-set expectations for ML performance and engineering performance

    • IT Services and IT Consulting
    • 1 - 100 Employee
    • Data Scientist
      • Jan 2016 - Dec 2016

      Combining internal proprietary customer data along with publicly available sources (Census, Economic Census, open government data, etc.), I built models for predicting consumer behaviour in rural Indian markets. • Model for early identification of defaulters in small business loans: predicted profits from using the model showed 100% improvement over existing approaches. Reduced costs by using only public datasets. • Model for predicting loan renewals in Self-Help Group microfinance… Show more Combining internal proprietary customer data along with publicly available sources (Census, Economic Census, open government data, etc.), I built models for predicting consumer behaviour in rural Indian markets. • Model for early identification of defaulters in small business loans: predicted profits from using the model showed 100% improvement over existing approaches. Reduced costs by using only public datasets. • Model for predicting loan renewals in Self-Help Group microfinance loans: built visual exploration tool for identify patterns across large sets of variables • Model for identifying ideal locations for small business loans Show less Combining internal proprietary customer data along with publicly available sources (Census, Economic Census, open government data, etc.), I built models for predicting consumer behaviour in rural Indian markets. • Model for early identification of defaulters in small business loans: predicted profits from using the model showed 100% improvement over existing approaches. Reduced costs by using only public datasets. • Model for predicting loan renewals in Self-Help Group microfinance… Show more Combining internal proprietary customer data along with publicly available sources (Census, Economic Census, open government data, etc.), I built models for predicting consumer behaviour in rural Indian markets. • Model for early identification of defaulters in small business loans: predicted profits from using the model showed 100% improvement over existing approaches. Reduced costs by using only public datasets. • Model for predicting loan renewals in Self-Help Group microfinance loans: built visual exploration tool for identify patterns across large sets of variables • Model for identifying ideal locations for small business loans Show less

    • United States
    • Technology, Information and Internet
    • 1 - 100 Employee
    • Data Scientist
      • Jul 2014 - Aug 2015

      Using person-level analytics data, I built models of online user behaviour for marketing optimisation and user experience. I worked with engineers to turn these prototypes into full-scale implementations. • Automated Conversion Funnels - discovers conversion paths that users are taking on websites • 'Similar People'​ Search - identify and cluster similar users based on their behaviour patterns • Conversion Optimiser - helped marketers identify actions causally linked to… Show more Using person-level analytics data, I built models of online user behaviour for marketing optimisation and user experience. I worked with engineers to turn these prototypes into full-scale implementations. • Automated Conversion Funnels - discovers conversion paths that users are taking on websites • 'Similar People'​ Search - identify and cluster similar users based on their behaviour patterns • Conversion Optimiser - helped marketers identify actions causally linked to conversions • Theme Analysis - identified interaction patterns across different customer segments Ad-hoc analyses to optimise Sales & Marketing business processes – • Analysed customers’ technology usage to identify partnerships & product integrations • Analysed marketing campaign data to assess the efficacy of individual campaigns • Profiled customers by plan type, vertical, company size, revenue, etc., for Sales • Helped Sales build a live demo using realistic customer data Show less Using person-level analytics data, I built models of online user behaviour for marketing optimisation and user experience. I worked with engineers to turn these prototypes into full-scale implementations. • Automated Conversion Funnels - discovers conversion paths that users are taking on websites • 'Similar People'​ Search - identify and cluster similar users based on their behaviour patterns • Conversion Optimiser - helped marketers identify actions causally linked to… Show more Using person-level analytics data, I built models of online user behaviour for marketing optimisation and user experience. I worked with engineers to turn these prototypes into full-scale implementations. • Automated Conversion Funnels - discovers conversion paths that users are taking on websites • 'Similar People'​ Search - identify and cluster similar users based on their behaviour patterns • Conversion Optimiser - helped marketers identify actions causally linked to conversions • Theme Analysis - identified interaction patterns across different customer segments Ad-hoc analyses to optimise Sales & Marketing business processes – • Analysed customers’ technology usage to identify partnerships & product integrations • Analysed marketing campaign data to assess the efficacy of individual campaigns • Profiled customers by plan type, vertical, company size, revenue, etc., for Sales • Helped Sales build a live demo using realistic customer data Show less

    • United States
    • Hospitals and Health Care
    • 1 - 100 Employee
    • Graduate Researcher - Psychological & Brain Sciences
      • Sep 2012 - May 2014

      I studied how humans navigate: potential applications include autonomous navigation, navigation guidance systems (such as Google Maps), urban design, etc. I designed experiments to study cue dependence & usage in human navigation. After collecting data from participants, I built an analysis pipeline in MATLAB to clean semi-structured data from log files, extract signatures of user behaviour, and communicated findings to both technical and non-technical audiences. Project: The Role… Show more I studied how humans navigate: potential applications include autonomous navigation, navigation guidance systems (such as Google Maps), urban design, etc. I designed experiments to study cue dependence & usage in human navigation. After collecting data from participants, I built an analysis pipeline in MATLAB to clean semi-structured data from log files, extract signatures of user behaviour, and communicated findings to both technical and non-technical audiences. Project: The Role of Orientation Cues in Human Navigation Studied effects of environmental cues on human navigation and decision-making • Built MATLAB pipeline to clean unstructured experiment data and extract navigation strategies • Designed quantitative metrics to capture dependence on cues • Identified hypotheses and designed experiments to test predictions • Presented poster at 17th International Conference of Cognitive and Neural Systems Teaching Assistant for 3 Psychology courses (250+ students) from Jan 2013 - May 2014 Show less I studied how humans navigate: potential applications include autonomous navigation, navigation guidance systems (such as Google Maps), urban design, etc. I designed experiments to study cue dependence & usage in human navigation. After collecting data from participants, I built an analysis pipeline in MATLAB to clean semi-structured data from log files, extract signatures of user behaviour, and communicated findings to both technical and non-technical audiences. Project: The Role… Show more I studied how humans navigate: potential applications include autonomous navigation, navigation guidance systems (such as Google Maps), urban design, etc. I designed experiments to study cue dependence & usage in human navigation. After collecting data from participants, I built an analysis pipeline in MATLAB to clean semi-structured data from log files, extract signatures of user behaviour, and communicated findings to both technical and non-technical audiences. Project: The Role of Orientation Cues in Human Navigation Studied effects of environmental cues on human navigation and decision-making • Built MATLAB pipeline to clean unstructured experiment data and extract navigation strategies • Designed quantitative metrics to capture dependence on cues • Identified hypotheses and designed experiments to test predictions • Presented poster at 17th International Conference of Cognitive and Neural Systems Teaching Assistant for 3 Psychology courses (250+ students) from Jan 2013 - May 2014 Show less

    • United States
    • Higher Education
    • 700 & Above Employee
    • Software Developer & Research Specialist
      • Nov 2010 - Jul 2012

      I built & maintained software that enabled neuroscience labs around the world to run experiments and collect data with extremely high temporal precision. I was solely responsible for identifying & resolving issues; development & testing; conducting benchmarking experiments, and publishing research findings. Project: MATLAB platform for experiments and recording data at sub-millisecond precision • Re-architected video & image presentation software to improve product usability &… Show more I built & maintained software that enabled neuroscience labs around the world to run experiments and collect data with extremely high temporal precision. I was solely responsible for identifying & resolving issues; development & testing; conducting benchmarking experiments, and publishing research findings. Project: MATLAB platform for experiments and recording data at sub-millisecond precision • Re-architected video & image presentation software to improve product usability & performance • Designed monitoring scripts and analyzed logs to measure software performance • Showed that new software beat previous timing benchmarks while handling complex stimuli • Demonstrated limitations of using MATLAB for collecting real-time behavioral data • Published findings on performance & usability improvements in Journal of Neurophysiology I also built an analytics and reporting tool to measure subject performance over time in experiments conducted in the lab. I also assisted researchers with various software-related tasks. Project: Custom Analytics & Reporting for Neuroscience Experiments • Designed scripts to collect daily data and monitor task progress • Analysed behavioural data and visualised KPIs • Used findings to devise optimised training schedules for improved task accuracy • Presented findings & recommendations to lab director and senior researchers Show less I built & maintained software that enabled neuroscience labs around the world to run experiments and collect data with extremely high temporal precision. I was solely responsible for identifying & resolving issues; development & testing; conducting benchmarking experiments, and publishing research findings. Project: MATLAB platform for experiments and recording data at sub-millisecond precision • Re-architected video & image presentation software to improve product usability &… Show more I built & maintained software that enabled neuroscience labs around the world to run experiments and collect data with extremely high temporal precision. I was solely responsible for identifying & resolving issues; development & testing; conducting benchmarking experiments, and publishing research findings. Project: MATLAB platform for experiments and recording data at sub-millisecond precision • Re-architected video & image presentation software to improve product usability & performance • Designed monitoring scripts and analyzed logs to measure software performance • Showed that new software beat previous timing benchmarks while handling complex stimuli • Demonstrated limitations of using MATLAB for collecting real-time behavioral data • Published findings on performance & usability improvements in Journal of Neurophysiology I also built an analytics and reporting tool to measure subject performance over time in experiments conducted in the lab. I also assisted researchers with various software-related tasks. Project: Custom Analytics & Reporting for Neuroscience Experiments • Designed scripts to collect daily data and monitor task progress • Analysed behavioural data and visualised KPIs • Used findings to devise optimised training schedules for improved task accuracy • Presented findings & recommendations to lab director and senior researchers Show less

    • United States
    • Software Development
    • 200 - 300 Employee
    • Program Manager
      • Jun 2009 - Nov 2010

      Studied malware threats to plan & release security features for Bing users • Received patent for inventing new method of detecting fake anti-virus software (‘scareware’) • Wrote scripts to identify malware-hosting websites from Bing’s search logs • Developed tools to protect users from malware targeting Internet Explorer vulnerabilities • Summarised research findings and new features in internal publication • Collaborated with researchers to identify new malware-detection… Show more Studied malware threats to plan & release security features for Bing users • Received patent for inventing new method of detecting fake anti-virus software (‘scareware’) • Wrote scripts to identify malware-hosting websites from Bing’s search logs • Developed tools to protect users from malware targeting Internet Explorer vulnerabilities • Summarised research findings and new features in internal publication • Collaborated with researchers to identify new malware-detection technologies Developed prototype for Bing.com real estate portal • Worked with providers to define real estate data acquisition protocol • Developed product specifications for real estate data processing pipeline • Designed mockups and user flow with designers for real estate vertical page • Troubleshot technical and design issues with software development team • Performed project management responsibilities to ensure that release targets were met Show less Studied malware threats to plan & release security features for Bing users • Received patent for inventing new method of detecting fake anti-virus software (‘scareware’) • Wrote scripts to identify malware-hosting websites from Bing’s search logs • Developed tools to protect users from malware targeting Internet Explorer vulnerabilities • Summarised research findings and new features in internal publication • Collaborated with researchers to identify new malware-detection… Show more Studied malware threats to plan & release security features for Bing users • Received patent for inventing new method of detecting fake anti-virus software (‘scareware’) • Wrote scripts to identify malware-hosting websites from Bing’s search logs • Developed tools to protect users from malware targeting Internet Explorer vulnerabilities • Summarised research findings and new features in internal publication • Collaborated with researchers to identify new malware-detection technologies Developed prototype for Bing.com real estate portal • Worked with providers to define real estate data acquisition protocol • Developed product specifications for real estate data processing pipeline • Designed mockups and user flow with designers for real estate vertical page • Troubleshot technical and design issues with software development team • Performed project management responsibilities to ensure that release targets were met Show less

    • Higher Education
    • 1 - 100 Employee
    • Undergraduate Researcher
      • May 2008 - Jul 2008

      Undergraduate Research Assistant • Investigated visual scene processing in the brain using fMRI • Performed Multivariate Pattern Analysis (MVPA) of fMRI data Research Programmer • Designed experiment for testing mechanism of change detection (using Psychophysics Toolbox) • Programmed probabilistic simulation for testing randomised data samples Undergraduate Research Assistant • Investigated visual scene processing in the brain using fMRI • Performed Multivariate Pattern Analysis (MVPA) of fMRI data Research Programmer • Designed experiment for testing mechanism of change detection (using Psychophysics Toolbox) • Programmed probabilistic simulation for testing randomised data samples

    • United States
    • Aviation and Aerospace Component Manufacturing
    • 100 - 200 Employee
    • Software Architecture & Integration Intern
      • May 2007 - Aug 2007

      Tested graphics drivers to low-level software defects. Developed testing criteria. Tested graphics drivers to low-level software defects. Developed testing criteria.

    • India
    • IT Services and IT Consulting
    • 700 & Above Employee
    • Intern
      • 2006 - 2006

      Defect-tested mobile phones for 3-Mobile, UK. Worked on video, audio and texting services. Defect-tested mobile phones for 3-Mobile, UK. Worked on video, audio and texting services.

Education

  • The Johns Hopkins University
    Master of Arts (MA), Psychological & Brain Sciences
    2012 - 2014
  • University of Illinois Urbana-Champaign
    BS, Computer Science
    2005 - 2009

Community

You need to have a working account to view this content. Click here to join now