Hello and welcome,

I'm a data science student — still early in my journey, but deeply curious and eager to grow. Here's a little of what I've been exploring and building so far.

~/portfolio
$ whoami
imbitnil

$ cat skills.json
{
  "foundation":    ["Math", "Stats", "CT"],
  "ai":    ["AI", "GenAI"],
  "ml":    ["MLF", "MLT", "MLP","MLOps"],
  "prog": ["PDSA", "DBMS",],
  "lang": ["Python", "Java", "SQL"],
  "lib":    ["NumPy", "Pandas", "Matplotlib"]
  "dev":    ["MAD I", "MAD II"]
  "se":    ["Software Engineering", "Software Testing"]
}

$

Featured Work

Research prototypes and production systems spanning detection, segmentation, 3D vision, and edge deployment.

NeuroSight

A real-time multi-object tracking and segmentation pipeline for autonomous driving. Fuses LiDAR point clouds with camera frames using a custom transformer-based architecture, running at 30 FPS on NVIDIA Orin.

PyTorch CUDA TensorRT ROS2 C++

DepthForge

Monocular depth estimation model achieving state-of-the-art results on NYU Depth v2. Uses a hybrid CNN-Transformer architecture with self-supervised pretraining on unlabeled video.

PyTorch Depth Estimation Transformers

SegAnything-Edge

Distilled and quantized version of SAM optimized for edge deployment. Runs interactive segmentation at 15 FPS on Jetson Nano with INT8 TensorRT inference.

ONNX TensorRT Jetson

NeuralRecon3D

Real-time 3D scene reconstruction from posed RGB images using neural implicit surfaces. Generates textured meshes from sparse views with NeRF-inspired volume rendering.

Python NeRF Open3D

AugmentKit

A high-performance data augmentation library for CV pipelines. GPU-accelerated transforms, mosaic augmentation, and domain-specific presets for medical, satellite, and autonomous driving data.

Python CUDA Albumentations

A bit about me

I am a data science student, currently learning how to understand data and make better, informed decisions. I am interested in exploring patterns, improving my analytical thinking, and building a strong foundation in data science.

My journey started with basic statistics and data analysis, and over time I began applying these concepts to real-world problems. I am still learning and continuously working on improving my understanding of data, models, and practical applications.

In my free time, I like studying new concepts, refining old concepts, working on small projects, and gradually improving my skills in data science.

Deep Learning

PyTorch / Lightning
TensorFlow / Keras
Hugging Face
Weights & Biases

Vision

OpenCV
Detectron2 / MMDet
Open3D / PCL
Albumentations

Languages

Python
C++ / CUDA
Rust
GLSL / Shaders

Deployment

TensorRT / ONNX
Triton Server
Docker / K8s
Jetson / Edge TPU

Experience

2024 - Present

Senior Computer Vision Engineer

Your Company

Leading the perception team building real-time multi-sensor fusion for autonomous systems. Designed a transformer-based detection pipeline achieving 45 mAP on internal benchmarks while maintaining 30 FPS on embedded hardware.

2022 - 2024

Computer Vision Engineer

Previous Company

Built and deployed instance segmentation models for industrial quality inspection. Reduced defect escape rate by 73% and optimized inference latency from 200ms to 35ms using TensorRT quantization.

2020 - 2022

ML Engineer / Research Assistant

University / First Company

Developed novel data augmentation strategies for medical image segmentation. Published 2 papers at MICCAI and contributed to open-source annotation tooling.

Research & Papers

Peer-reviewed publications at top computer vision and machine learning venues.

CVPR 2025

Efficient Multi-Scale Feature Fusion for Real-Time Panoptic Segmentation

qbit-glitch, J. Chen, A. Kumar, R. Patel

A novel lightweight fusion module that unifies semantic and instance segmentation branches, achieving 48.3 PQ on COCO Panoptic while running at 28 FPS on a single GPU.

ECCV 2024

Self-Supervised Depth Estimation via Cross-Modal Consistency in Dynamic Scenes

qbit-glitch, S. Li, M. Torres

A self-supervised framework that learns depth from monocular video by enforcing geometric consistency between predicted depth, optical flow, and ego-motion, with explicit handling of moving objects.

MICCAI 2023

Domain-Adaptive Augmentation Strategies for Low-Resource Medical Image Segmentation

qbit-glitch, K. Wang, D. Nakamura

Proposed a learned augmentation policy tailored to medical imaging that improves segmentation Dice score by 8.4% on polyp detection when training data is limited to 50 samples.

Writing & Experiments

Vision experiments, tutorials on model deployment, and deep dives into CV research.

Let's work together

Open to research collaborations, consulting on vision systems, or full-time roles in perception and computer vision.

Email GitHub LinkedIn X / Twitter Google Scholar