Menu

Hello There

I am Rohit Raju,
a generative AI
enthusiast.

Latest Projects More About Me
Scroll Down

About

More About Me

Howdy!

I am deeply passionate about machine learning and data science, constantly captivated by the potential of uncovering patterns within data and the countless opportunities it presents to make a positive impact on society. The sheer idea that simple mathematical principles can enable systems to detect these patterns has always enthralled me. My current research interests lies in the field of AI for education, specifically, intelligent tutoring systems. My focus lies on improving the current education system with the power of AI.

Inspired by my mother, a lifelong mathematics tutor whose philosophy is rooted in the belief that 'Every student has their own path, and it’s my responsibility to motivate and guide them through it', I have dedicated my academic and professional journey to creating personalized, AI-driven learning solutions. From my early days as a student assistant guiding peers through complex neural networks to developing tools that foster collaborative learning, my mission has always been clear: to make education more personal, inclusive, and impactful for diverse learners. This journey has taken me from creating viral YouTube content on effective learning to pioneering AI tools recognized at esteemed platforms like the UC Berkeley AI Hackathon, where my team ranked among the top ten for our narrative-based learning tool, "TaleTutor." Join me as I explore how AI-powered storytelling can unlock every learner’s potential, turning education into a truly unique and transformative experience, empowering students to rewrite the narrative of their own success.

I also find immense joy in playing the Tabla and immersing myself in the world of theater and drama. To support my interests, I work part-time as a YouTube content creator.

I've Got Some skills.

  • LEADERSHIP
  • COMMUNICATION
  • Presentation
  • RESEARCH
  • Large Language Modeling
  • MACHINE LEARNING
  • statistical data analysis
Download My CV
My Linkedin

My Work Experience.

Oct 2023-Present

Institute of Cognitive Science, CUB

Machine Learning Research Assistant,

Working on fine tuning LLM model built to improve collaborative skills among students in the National Science Foundation (NSF) AI Institute for Student-AI Teaming (iSAT), with the aim of transforming classrooms into more effective, engaging and equitable learning environments with the help of AI and HCI tools.

Nov 2022-Aug 2023

BOSCH

PLC Simulations System Designer (Internship)

Led the backend development of the SimuBridge simulation system, establishing connections between SimuBridge, OPC server, and DeviceBridge software for stress testing. Contributed to the conceptualization and layout of SimuBridge's user interface, optimizing design to facilitate efficient front-end coding for the team.

Dec 2022-Apr 2023

Infomatrix corporate solutions private limited

Backend Data Engineer (Internship)

Led the development of a reconciliation system's backend. Prioritized and sorted disordered data files and automated reconciliation, formerly a manual Excel process taking days, now achieves a remarkable 90% efficiency improvement. Overcame challenges, including the optimal rearrangement of completely disordered files, through strategic problem-solving.

Dec 2022-Apr 2023

Amrita School of Computing

Student Assistant to Dr. Peeta Basa Pati (Part time)

My contributions involved supporting the "Neural Networks and Deep Learning" course for B. Tech 3rd year students. I conducted an introductory TensorFlow session, supervised a 22-student volunteering project generating research data, evaluated their performance, and assisted fellow students with deep network language models for lab projects.

Portfolio

See My Latest Projects.

Here are some of the projects I have worked on!

YouTube Data Modeling

Machine Learning and LLM

Modeling YouTube title and thumbnail influence on views. Read the project here

Youtube Title Influence on Views: A Statistical Approach

Statistics

Testing the impact of Youtube title on views using statistical methods. View the project here

A System To Improve Noisy Text Accuracy Using LLMs

Large Languge Models

A System to Enchance Noisy Text documents using Deep Network Language Models. View the project here. View the paper here.

Auto Semantic & Syntactic Grader

Educational Technology

An evaluator to evaluate a code for it's smantic and syntatic structure. View the project here.

What People Say.

Author image

Mr. Rohit Raju served as a teaching assistant for my Neural Networks and Deep Learning course. In this role, he prepared and delivered a hands-on session on TensorFlow for the students in the class. This helped the students get started with their project implementation. Additionally, he managed the research volunteering contributions for a large group of students. He mentored two student groups for their course projects. I find Rohit to be confident, self-disciplined, committed as a leader with strong sense of time management and communication skills

Dr. Peeta Basa Pati Professor, Amrita University Bangalore
Author image

Rohit R seems Henri Frederic Amiel had you in mind when he said “ Doing easily what others find difficult is talent; doing what is impossible with talent is genius.”. Amrita Vishwa Vidyapeetham Amrita School of Engineering, Bangalore feels blessed to nurture talented genius students like Rohit R

Rashmi Verma Alumni Relations Office, Amrita University Bangalore

Publications

Journal & Conference Papers.

2023 IEEE I2CT Conference

System for Enhancing Accuracy of Noisy Text using Deep Network Language Models.

Text from image documents must be recognized for its usage. Various tasks such as plagiarism & error check, language analysis, information capture rely on the accuracy of this text conversion. OCR systems convert the document images to their text equivalent. These OCR systems are prone to introducing errors during the recognition process.This work reports a system developed to ingest image documents which is converted to text using available OCR technologies. The recognized text, subsequently, is processed with deep network language models to enhance the accuracy of text. The system consists of a client server architecture with user interface available from web application as well as from mobile app. For the language models, encoder-decoder based BART & MarianMT are used. The results obtained demonstrate a 35% reduction in WER using the BART language model.

Paper URL
2024 IEEE I2CT Conference

Comparative Study on Synthetic and Natural Error Analysis with BART & MarianMT.

Text is essential for communication, information sharing, knowledge acquisition, and analysis. It shapes public opinion, supports education, and drives online content, making it crucial in various domains. While there are various language models utilized for text analysis and text correction, there is little to no survey conducted on these model’s behavior and limitations. This work deals with studying BART and MarianMT language models behavior to an input dataset consisting of two types of errors, Synthetic and Natural. Synthetic errors are efficient to create and test, whereas Natural errors are more common and close to real world occurring errors. The models were trained and tested with the generated data, the results highlighted that BART exhibited consistent outputs towards both Synthetic and Natural errors and hence revealing a break-even point at the vicinity of 26% Synthetic error introduction. Conversely, the performance of MarianMT was comparatively diminished for Synthetic errors in contrast to Natural errors. These findings provide valuable insights into the behavior and capabilities of the models.

Paper URL
2024 JIKM Journal

Grammatical versus Spelling Error Correction: An Investigation into the Responsiveness of Transformer-Based Language Models Using BART and MarianMT

Text continues to remain a relevant form of representation for information. Text documents are created either in digital native platforms or through conversion of other media files such as images and speech. While the digital native text is invariably obtained through physical or virtual keyboards, technologies such as OCR & speech recognition are utilized to transform the images and speech signals to text content. All these variety of mechanisms of text generation also introduce error into the captured text. This project aims at analyzing different kinds of errors that occurs in text documents. The work employs two of the advanced deep neural network based language models, namely, BART and MarianMT, for rectifying the anomalies present in text. Transfer learning of these models with available dataset is performed to finetune their capacity for error correction. A comparative study is conducted to investigate the effectiveness of these models in handling each of the defined error categories. It is observed that while both the models are able to bring down the erroneous sentences by 20+%, BART is able to handle spelling errors far better (24.6%) than grammatical errors (8.8%).

Paper URL

Youtube Achievements.

Here are some of my minor achievements as a content creator

10000

Subscribers gained

1713004

Views gained

101

Videos uploaded

Contact

Say Hello.

Sending...
Your message was sent, thank you!
Your message was sent, thank you!

Email

Rohit.Raju@colorado.edu

Address

1600 Amphitheatre Parkway
Mountain View, CA
94043 US