Nachiketa Hebbar | AI Engineer | Researcher

Key Skills

Programming: Python, Java, SQL, C++
Frameworks & Tools: PyTorch, TensorFlow, Pytorch, Keras, Spark, Kafka
Cloud Platforms: AWS, GCP, Azure
Techniques: Machine Learning, Deep Learning, Generative AI, Predictive Modeling, NLP, Data Mining

About Me

Hi, I’m Nachiketa Hebbar, a Machine Learning Engineer 2 at TikTok, where I work on the Nearby Tab, building scalable recommendation systems and optimizing content ranking for local services in the U.S. I specialize in Recommendation Systems, Generative AI, and Deep Learning, and bring over two years of experience working with large language models, computer vision, and real-time ML systems.

I graduated from Carnegie Mellon University with a Master’s in Information Systems Management, and have previously worked at Awiros and Superkind on real-time computer vision and transformer-based recommendation systems. I also run a YouTube channel with over 25K subscribers and 2M+ views, where I teach Machine Learning and AI to a global audience.

As a researcher, I’ve authored 3 papers with over 100 citations, focusing on using Artificial Intelligence and Internet of Things to solve global challenges. I’m passionate about building impactful, AI-driven solutions that combine research, engineering, and product thinking.

Download My Resume

Machine Learning Projects

Here are some of the key projects I’ve worked on, spanning across different modalities in machine learning and based on productionizing AI systems.

Scalable Movie
Recommendation System

Built the entire MLOps pipeline to serve 1M+ user recommendations requests/week.

Details

Scalable Movie Recommendation System

End-to-end pipeline design for scalable movie recommendations.

Built scalable architecture on AWS.
Integrated with Kafka for real-time updates.

Skills: Python, PyTorch, Kafka, AWS

View Project

Text-Image
Search Engine

Built a web app to allow text-based image retrieval using CLIP and ChromaDB.

Details

Text-Image Search Engine

Leveraged CLIP embeddings for efficient text-image searches.

Implemented ChromaDB for vector indexing.
Developed a user-friendly web interface.

Skills: Python, PyTorch, CLIP, ChromaDB

View Project

Skin Cancer Detection

Springer published paper on detecting skin cancer using ensemble of ML and DL techniques.

Details

Skin Cancer Detection

Published paper exploring machine learning and deep learning techniques for cancer detection.

Used transfer learning for accuracy improvement.
Developed an ensemble of ML/DL models.

Skills: TensorFlow, Scikit-learn, Research Publication

View Paper

Deep Learning Model
Interpretability

Built a web app to compare and interpret image classification models.

Details

Deep Learning Model Interpretability

Developed tools to visualize model outputs for better interpretability.

Integrated Grad-CAM for model explanations.
Created an intuitive Gradio-based interface.

Skills: PyTorch, Gradio, Grad-CAM

View Project

Latent Diffusion Model
for 3D Metamaterials

Built a pipeline to accelerate synthetic metamaterials generation using LDMs.

Details

Latent Diffusion Model

Accelerated generation of 3D metamaterials using latent diffusion models.

Built custom pipelines for generative AI models.
Optimized inference for high throughput.

Skills: Python, Latent Diffusion, Generative AI

View Project

Sustainable AI Dashboard

Developed BNY Hackathon winning solution to forecast and reduce CO2 emissions of AI systems.

Details

Sustainable AI Dashboard

Forecasted carbon emissions of AI pipelines to optimize sustainability.

Built forecasting models for emission tracking.
Developed an interactive dashboard.

Skills: Python, Scikit-learn, Visualization

View Project

LLM-based Chatbot
Dispatcher Agent

Designed a dispatcher agent for routing users to the correct bot using LLMs.

Details

LLM Dispatcher Agent

Developed a bot routing system with fine-tuned LLMs for optimized query handling.

Used transformers for semantic query analysis.
Integrated with enterprise backend systems.

Skills: NLP, Transformers, Enterprise AI

View Project

Vehicle Attribute Classification

Combined object detection and image classification to identify attributes of vehicles moving at high speeds.

Details

Vehicle Attribute Classification

Deployed a computer vision application combining object detection and classification.

Processed high-speed traffic videos.
Integrated real-time classification models.

Skills: Python, PyTorch, Computer Vision

View Project

Research & Publications

I am passionate about advancing AI research, focusing on impactful solutions that address real-world problems. With 3 published papers and over 100 citations, my work spans machine learning, deep learning, and IoT applications.