Taniya Mathur

AI Researcher

Bengaluru, Karnataka, India6 yrs 8 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Expert in GenAI and Machine Learning with 6 years of experience.
Delivered innovative solutions in NLP and multimodal AI.
Proven track record in prompt engineering and data processing.

Stackforce AI infers this person is a Machine Learning and AI specialist with a focus on SaaS and Fintech solutions.

Contact

Skills

Core Skills

Large Language Models (llm)Retrieval-augmented Generation (rag)Prompt EngineeringMachine LearningNatural Language Processing (nlp)Data EngineeringComputer VisionWeb Development

Other Skills

A/B testingArtificial Intelligence (AI)Azure AI AssistantAzure AI StudioAzure OpenAICC++Cascading Style Sheets (CSS)CassandraData AnalysisData MiningData ScienceData StructuresData VisualizationDatabase Management Systems

About

GenAI and Machine Learning Specialist with ~6 years of experience in NLP, multimodal AI, and prompt engineering, delivering innovative, scalable solutions for real-world challenges.

Experience

Accenture

2 roles

AI/ML Computational Science Specialist

Promoted

Aug 2024 – Present · 1 yr 7 mos · Bengaluru, Karnataka, India · On-site

Contract Summarization & Document Q/A using LLMs - Delivered a GenAI solution for Telia NO B2B to automate contract document summarization and Q&A. Engineered and iteratively refined prompt designs for LLM-based contract summarization. Integrated Redis as a datastore for efficient query retrieval.
RAG Multimodal Knowledge Assistant - Built a Retrieval Augmented Generation (RAG) pipeline for Forvia to extract and process text, images, and tables from multiple document formats. Leveraged GPT-4 Turbo and LlamaIndex for multimodal data processing. Developed an Azure AI Studio Prompt Flow pipeline for seamless integration.
Smart Prompt - Designed and optimized AI-driven prompts for SAS script analysis using Azure AI Assistant for nationwide Economic Crime project, enhancing accuracy and efficiency in extracting meaningful insights and detecting errors. Led A/B testing by developing baseline and refined prompt variants, defining key metrics such as accuracy, completeness, and relevance. Measured performance through iterative evaluations, achieving a 95% accuracy rate in the refined version compared to 80% in the baseline.

Large Language Models (LLM)GPT-4Prompt EngineeringAzure OpenAIRetrieval-Augmented Generation (RAG)LLaMA+1

ML Engineering Specialist

Feb 2024 – Jul 2024 · 5 mos · Bengaluru, Karnataka, India · On-site

Sentiment Analysis Library Development - Designed and developed a Python-based library to classify email sentiments (Neutral, Negative, Positive). Implemented a Jenkins CI/CD pipeline for automated deployment. Integrated SonarQube for continuous code quality and security checks. Successfully deployed and maintained the solution for Barclays.

MLPythonJenkinsGitHubMachine Learning

Ey

Lead Analyst

May 2022 – Dec 2022 · 7 mos · Bengaluru, Karnataka, India · On-site

Hierarchical Category Prediction - Developed a text classification model to predict hierarchical category for products with description in Spanish language. Implemented a multi-step approach using word embeddings and model chaining to predict each level of the hierarchical category.
Dialogue Transcript Enhancement for Conversational AI - Implemented text data pre-processing and used NER to identify and extract entities from conversational data. Automated the rule-based approach to extract intent and call summaries via keyword-matching. This data was used for training
conversational bot.

Machine LearningNatural Language Processing (NLP)Scikit-LearnStatistical Data AnalysisPython (Programming Language)

Indium software

2 roles

Data Scientist

Aug 2021 – Mar 2022 · 7 mos · Remote

Data Engineering Pipeline for Credit-Analytics - Developed data pipelines aimed at seamless data migration from MySQL RDS to Cassandra, while standardizing data. Involved in data modelling for the Cassandra keyspace, ensuring optimal database performance. Built a data load pipeline, using PySpark, for the daily incremental updates, integrating data from the client APIs and transforming it into a standardized format for storage in Cassandra, significantly improving data update efficiency, with a ∼ 20% reduction in data processing time.

MySQLData EngineeringPySparkCassandra

Data Scientist

Aug 2020 – Aug 2021 · 1 yr · Remote

Fuel Optimization System for Maritime-based Clients - Designed ML model to provide real-time and near-accurate predictions for ship voyage parameters including Estimated Time of Arrival (ETA) and fuel consumption. The automated optimization recommendations helped optimize fuel throttle adjustments with ∼ 15% reduction in fuel consumption, reducing direct voyage costs by ∼ 20%, on an average, for our maritime client.

Machine Learning AlgorithmsExploratory Data AnalysisPython (Programming Language)Machine Learning

Aienterprise inc

2 roles

Data Scientist

Promoted

Apr 2019 – Apr 2020 · 1 yr · Bangalore · On-site

Search Query Classification - Led the design and development of a predictive model using Random Forest (RF) to categorize text queries into 87 predefined target categories. Utilized word2vec for word representation, including both the queries and product titles from clicked products. Conducted a rigorous evaluation of various tree-based approaches and finally selected the RF model, which achieved weighted F1 score of 0.91 on the test dataset.

MySQLMachine Learning AlgorithmsPythonStatistical Data AnalysisMachine Learning

Data Engineer

Oct 2018 – Apr 2019 · 6 mos · Bangalore · On-site

Face Recognition System - Designed and trained CNN based on Siamese architecture to learn face embeddings and map them to Euclidean space, where the distances represented measures of facial similarity. Implemented this system to identify and recognize human faces in real-time digital video feeds

Computer VisionPython (Programming Language)

Intel corporation

Software Engineer

Aug 2017 – Aug 2018 · 1 yr · Bangalore · On-site

Project Management Dashboard Enhancement:
Contributed to improving the project management dashboard by focusing on graph enhancements across multiple domains. Designed and implemented a dynamic front-end interface to facilitate program summary visualization within the dashboard. Additionally, integrated cron jobs for task scheduling and enabled efficient management of the hosting environment, including database administration. Further, customized the GUI to incorporate checklist functionality, enhancing the user experience for various programs within the dashboard

HTMLPHP FrameworksCascading Style Sheets (CSS)Web Development

Syncit services private limited

Summer Intern

May 2014 – Jul 2014 · 2 mos

Designed and implemented a website to provide real time information on availability of rooms in a hotel including automation of reservation and billing system, online room booking facility and booking confirmation sent to e-mail.