
Building AI for future
RAJ GUPTA.
Data Scientist & AI Engineer
Hi, I'm currently working as an AI Engineer and Researcher for Healthcare startup based out of middle east where I am building complex healthcare AI agents.
I have been working in AI for more than 2 years and have lead end to end Data Science and AI projects from AI architecture designing to deployment.
Proof of Work
A journey through my professional experiences and contributions

Doktor365
AI Engineer & Researcher (On Contract)
- Present
- Building AI agents for healthcare startup focused on middle east.
- AI architecture designing and development for simplifying healthcare operations and back office tasks.
- Leading multi agent system development and integration wih current CRMs for healthcare operations.
- Developing Tractable and precise algorithms to handle compliances and medical precisions.
Klaimz
Co-Founder & Data Scientist
Jan'24 - Oct'24
- Member of one of the 20 startups in India to be incubated at IIM Bengaluru NSRCEL under the Campus Founder Program.
- Designed entire Platform (mobile and web) for claims filing and analytics with a team of 6.
- Coded AI features to perform analytics task and query analysis.
- Worked on Sales, Marketing, Product Management, Managed finances, market research, business model canvas, and product market fit.
- We were not able to achieve the product market fit and had to shut down due to funded competitions.

Freelance Data Scientist
Independent Consultant
Oct'23 - Dec'23
- Worked with the FMCG client and built Process Mining and Analytics application.
- Developed data pipelines for collecting, processing, analyzing datasets and utilized large language model to provide analytics service.
Autodesk Inc
Data Scientist Intern
Jan'23 - July'23
- Built Internal Product: Recommend and connect with employees
- Created a end‐to‐end internal product which can recommend the employees to connect with based on given features such as skill and groups they like.
- Created multiple recommendation algorithms using AWS Personalise, Statistics and AWS Sagemaker, which were able to provide excellent results.
- Performed the data science pipeline on my own, from feature engineering to creating the User Interface for showcasing to stakeholders.
- Optimized the algorithm further by 40% and created multiple filters for recommendations.
- Built Internal project: Time Series predictive analysis and forecasting algorithm
- Created a end‐to‐end algorithm to help stakeholder by provide forecasting based on occupancy sensors data for offices at different geographic locations.
- The project was about defining the problem statement and performing Feature engineering.
- Performed multiple analysis to come up with a proper feature that can optimize the workspace utilization.
- Developed and tested multiple forecasting models and performed hypothesis testing, created our own feature for forecasting and built an User Interface for showcasing it to stakeholders.
- This project provides positive financial impact for Autodesk as it reduce the operating cost and optimize the utilization of space.

MetabrixLabs
Computer Vision Intern
Oct'22 - Dec'22
- Researched on creating 3D and 4D avatars for faces and implementing papers like neural head avatars and neural bodies and implementation of 2D images into 3D objects.
- Implemented methods for adding textures over human mesh using 2D photos with diffusion models.

IIT Kharagpur, Dept. of Computer Science
AI Research Intern
May'22 - July'22
- Worked on research paper "Detection of Disease using Volunteered Geographical Information from Twitter Using Bidirectional LSTM and Pretrained BERT" under Dr. Pabitra Mitra.
- Collected the data from Twitter around 2000 tweets bases on hashtags with disease names like COVID-19, tuberculosis, and malaria. The tweets were manually annotated for classification tasks and sentiment analysis work.
- Executed the Data Science pipeline for the research paper, encompassing raw data collection from Twitter, data annotation, model development, and performance evaluation.
- My research presented a method for detecting epidemic and with better quality of data and newer SOTA algorithms it is feasible and faster to detect epidemics when we involve Volunteered information from social medias.

CloudyML
Data Science course creator
Jan'22 - Mar'22
- Created courses on Computer Vision and Machine Learning. The course is enrolled by hundreds of learners to build foundational knowledge.

Skills & Expertise
Specialized in AI, Machine Learning, and Data Science
- Data Science & ML
- ·
- Statistics & Python
- ·
- AWS & Cloud
- ·
- ML Libraries
- ·
- Time Series
- ·
- Neural Networks
- ·
- Vision & NLP
- ·
- Computer Vision
- ·
- Object Detection
- ·
- Image Processing
- ·
- NLP & Transformers
- ·
- LSTM & RNN
- ·
- Analytics & Development
- ·
- Data Analytics
- ·
- Tableau
- ·
- Business Analytics
- ·
- Data Visualization
- ·
- Software Development
- ·