Gaurav T.

AI Researcher

Bengaluru, Karnataka, India11 yrs 1 mo experience

Most Likely To SwitchHighly Stable

Key Highlights

Expert in Generative AI and Multi-Agent Applications.
Proficient in fine-tuning Large Language Models.
Strong mentor for junior AI and ML engineers.

Stackforce AI infers this person is a Data Science and AI specialist with a focus on Healthcare and Telecom industries.

Contact

Skills

Core Skills

Generative AiMachine LearningDeep LearningData ScienceNatural Language ProcessingData Analysis

Other Skills

AWSAWS SageMakerAWS TextractAlgorithmsAnacondaAnalytical SkillsAnalyticsApplication Programming Interfaces (API)Artificial Intelligence (AI)Artificial Neural NetworksAzure DatabricksBERT (Language Model)BERT modelsClassification AlgorithmsCluster Analysis

About

As a Principal AI/ML Engineer, I am dedicated to advancing the field of artificial intelligence by leveraging state-of-the-art Generative AI (GenAI) technologies, including embedding models and Multi-Modal Retrieval-Augmented Generation (RAG). My current focus involves the development of GenAI agents and multi-agent applications using cutting-edge tools like LangGraph, LangChain, Llama Index, and CrewAI. Additionally, I work on applications based on the OpenAI Realtime API, ensuring seamless integration of generative models for real-time solutions that meet diverse business needs. In conjunction with my work in GenAI, I utilize Azure AI to deploy OpenAI models for GenAI-based product applications, crafting innovative solutions that address real-world challenges. My expertise spans a diverse range of technologies, including Machine Learning Algorithms, Neural Networks, and Statistical Analysis. I am proficient in Python, Matlab, and essential libraries such as Pandas, NumPy, Scikit-learn, and SciPy, equipping me to tackle complex data challenges and extract meaningful insights. With hands-on experience in frameworks like TensorFlow, PyTorch, H2O, and Hugging Face, I specialize in Large Language Model fine-tuning, Multimodal Models, CNNs, Transformers, and Prompt Engineering. I am also experienced in fine-tuning LLMs using techniques like LoRA and QLoRA. In addition to my technical skills, I am passionate about mentorship and guiding junior colleagues through their projects, fostering a collaborative environment to help them grow in AI and machine learning.

Experience

11 yrs 1 mo

Total Experience

2 yrs 2 mos

Average Tenure

3 yrs 4 mos

Current Experience

Chubb

3 roles

Principal AI/ML Engineer

Promoted

Apr 2025 – Present · 1 yr 2 mos

As a Principal AI/ML Engineer at Chubb, my role encompasses the application of advanced technologies such as Generative AI (GenAI), Multi-Agent Applications, and embedding models, alongside Multi-Modal Retrieval-Augmented Generation (RAG). I work extensively with tools like LangGraph, LangChain, Llama Index, and CrewAI to develop innovative solutions and scalable applications utilizing the OpenAI Realtime API.
My work involves designing and developing sophisticated AI solutions to address complex business challenges. I employ state-of-the-art techniques, including both traditional machine learning (ML) algorithms and deep learning (DL) models, for natural language processing, computer vision, and speech applications. My experience spans a variety of ML approaches, such as regression, classification, clustering, and decision trees, as well as advanced DL techniques utilizing CNNs, RNNs, and Transformers. Additionally, I specialize in fine-tuning Large Language Models (LLMs) and excel in prompt engineering, creating and optimizing prompts to enhance the performance of LLMs and transformers.
I have successfully deployed AI models at scale, ensuring their seamless integration with existing business workflows while maintaining performance and reliability. Proficient in Python, I utilize it extensively to build end-to-end AI and machine learning solutions.
In my capacity as a mentor, I guide junior team members to help them enhance their skills and deepen their understanding of both traditional and advanced machine learning techniques. I also emphasize best practices for project management and foster a collaborative team environment.

Generative AIMulti-Agent Applicationsembedding modelsMulti-Modal Retrieval-Augmented GenerationLangGraphLangChain+5

Lead AI/ML Engineer

Oct 2024 – Apr 2025 · 6 mos

As a Lead AI/ML Engineer at Chubb, my role encompasses the application of advanced technologies such as Generative AI (GenAI), Multi-Agent Applications, and embedding models, alongside Multi-Modal Retrieval-Augmented Generation (RAG). I work extensively with tools like LangGraph, LangChain, Llama Index, and CrewAI to develop innovative solutions and scalable applications utilizing the OpenAI Realtime API.
My work involves designing and developing sophisticated AI solutions to address complex business challenges. I employ state-of-the-art techniques, including both traditional machine learning (ML) algorithms and deep learning (DL) models, for natural language processing, computer vision, and speech applications. My experience spans a variety of ML approaches, such as regression, classification, clustering, and decision trees, as well as advanced DL techniques utilizing CNNs, RNNs, and Transformers. Additionally, I specialize in fine-tuning Large Language Models (LLMs) and excel in prompt engineering, creating and optimizing prompts to enhance the performance of LLMs and transformers.
I have successfully deployed AI models at scale, ensuring their seamless integration with existing business workflows while maintaining performance and reliability. Proficient in Python, I utilize it extensively to build end-to-end AI and machine learning solutions.
In my capacity as a mentor, I guide junior team members to help them enhance their skills and deepen their understanding of both traditional and advanced machine learning techniques. I also emphasize best practices for project management and foster a collaborative team environment.

Generative AIMulti-Agent Applicationsembedding modelsMulti-Modal Retrieval-Augmented GenerationLangGraphLangChain+5

Lead Data Scientist

Feb 2023 – Jan 2025 · 1 yr 11 mos

As a Lead Data Scientist at Chubb, my role includes working on advanced technologies such as Deep Learning Algorithms, Transformers, MultiModal Models, Prompt Engineering, Building Applications using LLMs, Deploying and Finetuning LLMs, and working with Python. Additionally, I am responsible for mentoring junior team members.
My work involves designing and developing sophisticated machine learning solutions to solve complex business problems. I apply state-of-the-art techniques to develop new models for natural language processing, computer vision, and speech applications. I have experience in model training and testing using a wide range of statistical and deep learning techniques. I also excel in prompt engineering, where I create and fine-tune prompts to improve the LLMs and transformers.
I have successfully deployed models at scale, ensuring their smooth integration with existing business workflows. I am proficient in Python programming and have used it extensively to develop end-to-end machine learning solutions.
In my role as a mentor, I guide junior team members to help them grow their skills and develop their abilities in machine learning techniques, while also teaching them best practices for project management and teamwork.

Deep Learning AlgorithmsTransformersMultiModal ModelsPrompt EngineeringPythonMachine Learning+1

Aarete

Lead Data Scientist

Sep 2022 – Jan 2023 · 4 mos · Pune, Maharashtra, India

As a data scientist, I utilized algorithmic and programming tools to build predictive models and established a quantitative framework for evaluating and analyzing alternative techniques. My work involved:
Utilizing statistical and data mining techniques, such as hypothesis testing, machine learning, and retrieval processes, on large data sets to identify trends, figures, and other relevant information.
Using time series data to build various regression and classification algorithms, such as Linear Regression, Decision Trees, Ensemble techniques, SVM, Bagging, Boosting, GBM, etc., by employing various sk-learn libraries.
In my previous projects, I worked on two key areas:
1. Automatically Reading Information from Healthcare Contracts using NLP:
I developed algorithms to read contracts automatically and extract vital information into a JSON file for further analysis. I used various BERT models like roBERTa, XLM-roBERTa, and Simple Transformers for building the application. Additionally, I also tried to use AWS comprehend for the same purpose. AWS Textract was used to read entities from healthcare contracts. Various NER models were also used for detecting entities from contract like Effective date and Provider Names.
2. Outlier Detection in Healthcare Claims Data:
I employed various statistical and visualization techniques, such as IQR and Local Outlier Factor, OneClass SVM, Isolation Forest for detecting outliers in claims data. Outliers were then analyzed to detect if claims were wrongly booked for any recoupment. I ensured the successful deployment of the solution on AWS Pipeline.

statistical techniquesdata mining techniquesregression algorithmsclassification algorithmsBERT modelsAWS+2

Rolls-royce

Data Scientist

Apr 2019 – Sep 2022 · 3 yrs 5 mos · India · On-site

In my role as a data scientist, I utilized algorithmic and programming tools to build helpful predictive models. My work included:
Utilizing exploratory data analysis techniques to identify patterns, relationships, and trends.
Creating reports and statistical analysis as required and working with stakeholders to identify opportunities for growth.
Assessing the effectiveness and accuracy of new data sources and gathering techniques.
Providing a quantitative framework for evaluating and analyzing alternative techniques.
Working with statistical and data mining techniques, such as hypothesis testing, machine learning, and retrieval processes, on large data sets to identify trends, figures, and other relevant information.
Developing advanced algorithms that solve problems of large dimensionality in a computationally efficient and statistically effective manner.
Working with Time Series Data, building various regression and classification algorithms by using various sk-learn libraries such as Linear Regression, Decision Trees, Ensemble techniques, and SVM, Bagging, Boosting, GBM.
I have worked on several key projects, including:
1. Injector Remaining Useful Life Prediction:
2. Sensor Anomaly Detection using Deep Learning and Auto Encoders:
3. Turbocharger Load Cycle Analysis and Failure Prediction:
4. Spark Plug Remaining Useful Life Calculation:

predictive modelsexploratory data analysisregression algorithmsclassification algorithmsData ScienceMachine Learning

Infosys

Data Scientist

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

In this role, I worked closely with business and engineering teams to encourage statistical best practices with respect to experimental design, data capture, and data analysis. My work included:
Working on ML Project Prospect Modeling: This project involved creating prospect models for a Telecom client in Belgium. I built classification models (Logistic, Random Forest, XGBoost) to cross-sell/up-sell the products. I also developed docile charts to determine the target customers for marketing and worked on BI tools such as Tableau for data visualizations to enhance business insights and facilitate the decision-making process. Furthermore, I provided analytic support to other team members with predictive modeling and statistical techniques.
I also successfully completed the following project:
1. Customer Segmentation for Telecom Services:
I classified customers into different segments for targeted advertisement and identified various features from datasets which were useful for the classifying customers. I calculated the scores for each feature specifying how important it is to effectively segment customers into different classes.

classification modelspredictive modelingdata visualizationData ScienceMachine Learning

Ibm

Data Analyst

Apr 2015 – Oct 2018 · 3 yrs 6 mos · Bengaluru Area, India · On-site

In my this role, I directed the development, validation, testing, and implementation of analytical products and applications. My work included:
Working on an analytics project to classify support tickets created at service centers according to the problem. I worked with various statistical techniques such as ANOVA, Factor Analysis, Cluster Analysis, Logistic Regression, Decision Trees, Random Forest to enhance the project work.
Furthermore, I explored and examined data from multiple disparate sources and extracted, transformed, and combined all incoming data with the goal of discovering a previously hidden insight. I was responsible for data analytics to drive decision-making and lean on automation and machine learning as core. I also was responsible for data visualization using different plots and graphs.
I successfully completed the following projects:
1. CHNi Service Ticket Classification:
I worked on developing a tool used to classify the service tickets generated by service engineers into different classes. I used NLP and Text analytics to generate features from the Tickets. I used the clustering to cluster different tickets into different zones. Then, I worked with the Business team to label the tickets. After labeling, I used the datasets to build ML models Tree-based to classify new tickets into respective classes.
2. CHNi Account Management:
I collected and analyzed the details about the hours spent by different teams for the CHNi account. I created dashboards for effective tracking of productivity for different teams within the CHNi account. I also created a prediction for future quarters of utilization.

statistical techniquesdata visualizationdata analyticsData AnalysisMachine Learning