Sudha M. profile photo 1 on Tal

Sudha M.

Senior Data Scientist | GenAI | NLP | ML Systems | IIT Madras

Bengaluru

9+ years of work experience

Azent Overseas Education
Oct 20205 yrs 11 mosPresent

Mumbai, Maharashtra, India

Recommendation System : Automated the process of course recommendation to the study abroad aspirants using quantitative and supervised methods. Reduced the manual effort and time taken to recommend by 85% Azent Chatbot : Built a Whatsapp bot which answers user queries about study abroad space and helps users resolve their complaints Marks Extractor : Worked on extracting name, DOB, board name, overall GPA, respective marks/grades in individual subjects etc from school and university transcripts Fairs Automation : Built the pipeline to recommend suitable universities to students attending the educational fairs. Drastically reduced the pre-counselling time Passport Details Extractor : Extracted the name, DOB, place of birth etc details present in the passport to enable auto filling Programmatic SEO : Automated the process of creating landing pages for different countries, programs, centres, universities Status Updatinator POC : Worked on a service which updates the student's application status periodically in CRM. It fetches the email received from the partner/university, processes it and extracts necessary info, updates the status/remarks in the CRM and noties the respective student counsellor about it. LOR POC : Worked on automating the process of creating Letters of recommendation for students.
Netomi
Dec 20181 yr 6 mosMay 2020

Gurgaon, Haryana

Intent Classification : Improved the Accuracy and Coverage of Intent classifier by employing BERT based fine tuned models. Email Signature Detection : Improved over the rule based baseline model by using various Deep Learning methods like Autoencoder based approach, Start and End token Detection, Semantic Textual Similarity. Experimented on representing chat data to detect the intent based on customer's response. Worked on clustering untagged data to increase Fragmentation and Coherence
Genesys
Aug 20184 mosNov 2018

Hyderabad, Telangana

Increased the PII masking tool’s performance by automating various dataset correction techniques. Expanded the practical utility of the tool by performing better on Date and Time related entities. Worked on building an automatic agent response tool on a fixed domain, using Supervised learning, Online learning.
IIIT Hyderabad
Jan 20171 yr 8 mosAug 2018

Explored various applications and research directions in NLP. Implemented models for various tasks like Summarization, Question Answering and Pun Detection.
AgroStar
Feb 20164 mosMay 2016

Pune, Maharashtra

Worked on building Purchase Order Module for Agrostar e-commerce platform. Included developing Vendor Onboarding, Management Module, Integration between PO system and Warehouse Management system.

Sudha’s AI Stack

codex
codex
hugging face
hugging face
langchain
langchain
langfuse
langfuse
langgraph
langgraph
openai
openai

Education

Indian Institute of Technology, Madras

Bachelor's degree, Computer Science

Indian Institute of Technology, Madras

Join Sudha and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Sudha M. on Tal

Headline

Senior Data Scientist | GenAI | NLP | ML Systems | IIT Madras

About

As a Senior Data Scientist at Azent Overseas Education, I apply my expertise in natural language processing (NLP) and recommendation systems to automate and enhance the process of course and university selection for study abroad aspirants. I have built and deployed scalable solutions that reduce manual effort and time, increase user retention and satisfaction, and improve business outcomes. I have over five years of industry experience in data science, working on diverse and challenging problems such as intent classification, email signature detection, question answering, and chatbot development. I have also contributed to research projects on PII masking, automatic agent response, and online learning. I hold a Bachelor of Technology degree in Computer Science from Indian Institute of Technology, Madras. I am passionate about applying data science to solve real-world problems and create positive impact.

Location

Bengaluru

AI stack

codex

hugging face

langchain

langfuse

langgraph

openai

Work experience

Senior Data Scientist at Azent Overseas Education, Oct 2020 - Present · 5 yrs 11 mos, Mumbai, Maharashtra, India, Recommendation System : Automated the process of course recommendation to the study abroad aspirants using quantitative and supervised methods. Reduced the manual effort and time taken to recommend by 85% Azent Chatbot : Built a Whatsapp bot which answers user queries about study abroad space and helps users resolve their complaints Marks Extractor : Worked on extracting name, DOB, board name, overall GPA, respective marks/grades in individual subjects etc from school and university transcripts Fairs Automation : Built the pipeline to recommend suitable universities to students attending the educational fairs. Drastically reduced the pre-counselling time Passport Details Extractor : Extracted the name, DOB, place of birth etc details present in the passport to enable auto filling Programmatic SEO : Automated the process of creating landing pages for different countries, programs, centres, universities Status Updatinator POC : Worked on a service which updates the student's application status periodically in CRM. It fetches the email received from the partner/university, processes it and extracts necessary info, updates the status/remarks in the CRM and noties the respective student counsellor about it. LOR POC : Worked on automating the process of creating Letters of recommendation for students.

Data Scientist at Netomi, Dec 2018 - May 2020 · 1 yr 6 mos, Gurgaon, Haryana, Intent Classification : Improved the Accuracy and Coverage of Intent classifier by employing BERT based fine tuned models. Email Signature Detection : Improved over the rule based baseline model by using various Deep Learning methods like Autoencoder based approach, Start and End token Detection, Semantic Textual Similarity. Experimented on representing chat data to detect the intent based on customer's response. Worked on clustering untagged data to increase Fragmentation and Coherence

Research Intern at Genesys, Aug 2018 - Nov 2018 · 4 mos, Hyderabad, Telangana, Increased the PII masking tool’s performance by automating various dataset correction techniques. Expanded the practical utility of the tool by performing better on Date and Time related entities. Worked on building an automatic agent response tool on a fixed domain, using Supervised learning, Online learning.

Research Intern at IIIT Hyderabad, Jan 2017 - Aug 2018 · 1 yr 8 mos, Explored various applications and research directions in NLP. Implemented models for various tasks like Summarization, Question Answering and Pun Detection.

Software Developer at AgroStar, Feb 2016 - May 2016 · 4 mos, Pune, Maharashtra, Worked on building Purchase Order Module for Agrostar e-commerce platform. Included developing Vendor Onboarding, Management Module, Integration between PO system and Warehouse Management system.

Education

Bachelor's degree, Computer Science, Indian Institute of Technology, Madras