Engineering AI systems for scale and impact.

About

I design and deploy machine learning architectures that operate reliably in demanding environments. My focus bridges generative modeling, computer vision, and robust data infrastructure—taking complex requirements from initial architecture through to production scale.

I prioritize system resilience, latency optimization, and observable metrics just as highly as algorithmic performance. Leading engineering efforts means I work closely with cross-functional teams to ensure the solutions we build directly address core business challenges. Day to day, I translate ambiguous problems into structured technical roadmaps and deliver systems that drive measurable outcomes.

Experience

Chief AI Engineer, Plexor

Oct 2025 to Present

  • Direct the architecture and deployment of real-time object detection and vision-language models for retail surveillance.
  • Design fault-tolerant edge AI pipelines that operate reliably in constrained store environments.
  • Push the boundaries of detection precision, establishing new benchmarks for the platform's vision capabilities.

Junior Specialist, University of California, Riverside

Jul 2024 to May 2025

  • Developed deep learning classifiers for mosquito orientation, reaching 96% accuracy.
  • Built robust image processing workflows utilizing thresholding and edge detection to isolate critical biological features.
  • Streamlined deployment pipelines, enabling interdisciplinary teams to scale feature extraction for broader studies.

Research Engineer, University of Guelph

Jan 2023 to Jun 2023

  • Combined image segmentation with LLM-driven text analysis to evaluate soil properties, yielding an 18% accuracy gain.
  • Engineered automated tooling that cut data processing time by 40% and improved indicator precision.
  • Partnered directly with domain experts to fine-tune model heuristics for real-world agricultural use cases.

Computer Vision Researcher, Avawatz, UT Dallas, IIT Bombay

Mar 2022 to Jul 2022

  • Trained object detection algorithms for autonomous navigation using Faster R-CNN and YOLO, driving a 15% accuracy improvement.
  • Leveraged 3D simulations to boost segmentation precision by 20% for off-road environments.
  • Optimized preprocessing pipelines to reduce large dataset training times by 25%.

Machine Learning Engineer, Lokavidya, IIT Bombay

Jan 2021 to Sep 2022

  • Architected NLP-driven knowledge curation engines, significantly improving search relevance and retrieval speed.
  • Built and tuned classification and clustering models to structure chaotic datasets, minimizing data redundancy.
  • Automated anomaly detection pipelines, cutting manual verification overhead by 20%.

Skills

Core Capabilities
Deep Learning
NLP
LLMs
Computer Vision
Generative Modeling
Diffusion Models
Machine Learning
Reinforcement Learning
Data Analysis
Data Visualization
Statistics
Graph Neural Networks
Languages & Frameworks
Python
C++
C
Keras
TensorFlow
PyTorch
PyTorch Lightning
FastAI
OpenCV
Numpy
Matplotlib
Pandas
Hugging Face
SQL
Flask
BeautifulSoup
XGBoost
Tableau
MLOps
GCP
AWS
Kubernetes
Docker
Git CLI
Jupyter Notebooks
Anaconda
Linux Terminal
ROS
CUDA
MPS (Apple)
Apache Spark
Tensorboard
Firebase
MS Azure
MATLAB

Projects

ApollodB: Psychoacoustic EQ Generation System

SOTA Project - 2024 to Present
  • Architected a TensorFlow-based psychoacoustic EQ generator, deployed on Google Cloud, achieving state-of-the-art metrics on the DEAM dataset.
  • Engineered custom neural networks that output highly personalized audio profiles grounded in psychoacoustic principles.
  • Open-sourced the complete platform, including the interactive validation suite, for community adoption and iteration.
Python TensorFlow Google Cloud Psychoacoustics DEAM

Generative Networks and VAEs to Assist Artists

SRM IST - Jan 2023 to Jun 2023
  • Developed Variational Autoencoders (VAEs) for sketch refinement, demonstrating a 30% improvement in artist workflow efficiency.
  • Hit 90.6% accuracy on the CUHK Face Sketch Database; presented findings at ICCET 2023.
  • Designed the architecture specifically to preserve high-frequency details, closing the gap between generative output and practical artistic needs.
Python Keras TensorFlow OpenCV VAE

Energy Price Forecasting Using LSTMs and ARIMA

SATU - 2022
  • Built hybrid LSTM/ARIMA forecasting pipelines, lifting prediction accuracy by 18% on complex Australian energy datasets.
  • Integrated NLP-driven trend analysis to surface broader market context for key stakeholders.
  • Presented the methodology and scalability trade-offs at the AIJR Conference 2022.
Python LSTM ARIMA NLP
PersonaForge: Fine-Tuned LLM Assistant - UC Riverside
Dec 2024
  • Fine-tuned an offline personal LLM using llama.cpp to handle day-to-day task automation contextually.
  • Optimized for local-only deployment to guarantee strict data privacy constraints without sacrificing response latency.
Python LLM llama.cpp Personal Assistant NLP
FormScribe: AI-Powered Web Form Automation
Apr 2025
  • Automated complex data entry workflows by bridging Selenium web drivers with local LLaMA 2 7B models.
  • Designed the architecture to function entirely offline, prioritizing robustness and data security across varied browser environments.
Python Selenium Ollama LLaMA 2 7B AI Automation

Education

Master of Science, Computer Science, UC Riverside

Sep 2023 to Dec 2024

Bachelor of Technology, Computer Science, SRM IST

2019 to 05/2023