# Omer Tariq

Omer Tariq is a remote worker on Himalayas, AI Research Scientist at Neubility, based in South Korea.

Source: https://himalayas.app/@omertariq2

```text
AI Research Scientist at Neubility building 25FPS multi-camera perception for autonomous delivery robots.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | South Korea |
| Authorized to work in | South Korea |
| Salary expectations | 60020k+ KRW |
| Job categories | AI Research Scientist, Machine Learning Engineer, Research Scientist |
| Skills | 4D Perception, Embodied AI, Vision Language Models, Multi Camera Perception, 3D Detection, Depth Estimation, Open Vocabulary Detection, Semantic Segmentation, Model Quantization, Knowledge Distillation, LoRAs, Model Compression, Vision Language Action (VLA) Policies, Visual Inertial SLAM, Camera–LiDAR–IMU Fusion |
| Website | https://omertariq-kaist.github.io |
| LinkedIn | https://linkedin.com/in/omer-tariq-k2021 |
| GitHub | https://github.com/OmerTariq-KAIST |

## About

```text
At Neubility, I lead 3D/4D scene perception for autonomous urban delivery robots. I developed VPE-Neubie, a multi-camera pipeline that unifies detection, depth estimation, open-vocabulary recognition, and segmentation, running at 25FPS on a Jetson Orin.

I also proposed GAPT, a post-training quantization framework that protects grounding-critical pathways in vision–language models, and developed SAD-LoRA for distillation-based compression. I prototyped long-horizon vision–language–action policies and validated them in NVIDIA Isaac Sim and CARLA.

At KAIST, I proposed domain-invariant inertial localization architectures and developed privacy-preserving federated learning systems. My work spans embodied AI, efficient multimodal models, robotics, and edge deployment.
```

## Experience

### AI Research Scientist (current)

Neubility. Aug 2025 - Present

```text
Leading 3D/4D scene perception for autonomous urban delivery robots, developing a production multi-camera pipeline unifying 3D object detection, monocular depth estimation, open-vocabulary recognition, and semantic segmentation. Proposed quantization and distillation frameworks for vision-language models and prototyped long-horizon vision-language-action policies.
```

Skills: 4D-Perception, Embodied-AI, Video-Understanding, DINOv3, Mixed-Precision, Jetson-Orin, GAPT, SAD-LoRA, Anti-Shortcut-Distillation, Isaac-Sim, CARLA, Flow-Matching, Vision-Language-Action-(VLA)-Foundation-Models

### Machine Learning Engineer

Monitra. Feb 2025 - Sep 2025

```text
Improved model robustness by 15% and inference throughput by 8x using diffusion-based data augmentation, self-supervised pretraining, and PyTorch-TensorRT optimization. Deployed a real-time PRPD fault-detection service on Azure Kubernetes Service with an MLOps stack.
```

Skills: Computer-Vision, ML-Systems, Diffusion-Models, Self-Supervised-Learning, PyTorch, TensorRT, Azure-Kubernetes-Service, fastAPI, Kafka, MLFlow, GitHub-Actions, CI-CD

### Ph.D. Candidate

Intelligent Service Integration Lab, KAIST. Aug 2021 - Jun 2025

```text
Proposed domain-invariant architectures for inertial localization, a hardware-aware multiscale transformer for TinyML, and adapted BEVFormer for embedded multi-camera 3D detection. Fused multi-sensor streams and developed privacy-preserving federated learning systems for on-device odometry.
```

Skills: DeepILS, ConvXformer, NanoMST, BEVFormer, Sensor-Fusion, Visual-Inertial-SLAM, FPGA, Federated-Learning, Differential-Privacy, TinyML, Quantization

### Research Intern

DAIM Lab, KAIST. Nov 2023 - Mar 2024

```text
Integrated LiDAR, IMU, and wheel-odometry SLAM for an industrial autonomous mobile robot on power-optimized embedded hardware, meeting a warehouse positioning target of less than 5cm error.
```

Skills: LiDAR, IMU, Wheel-Odometry, Simultaneous-Localization-and-Mapping-(SLAM), Embedded-Systems, Sensor-Fusion

### Senior Manager, SoC & AI Accelerators

National Electronics Complex (NECOP). Apr 2019 - May 2022

```text
Led SoC/RTL verification and FPGA development for embedded AI across four production programs, contributing to two first-pass silicon tape-outs.
```

Skills: SOC, RTL-Verification, FPGA, Embedded-AI, Tape-Out

### Assistant Manager, FPGA & Signal Processing

National Space Agency (SUPARCO). Oct 2014 - Apr 2019

```text
Designed radiation-tolerant FPGA-based DSP systems for satellite communication and imaging payloads sustaining over 99.9% uptime in orbit; recognized with a Distinguished Service Award for the PakTes-1A launch.
```

Skills: FPGA, DSPs, Radiation-Tolerant-Design, Satellite-Communications, Signal-Processing

## Education

### Korea Advanced Institute of Science & Technology (KAIST)

Ph.D., Computer Science. 2021 - 2025. Grade: 3.73/4.3

Activities and societies: KAIST MS/Ph.D. Scholarship

```text
Ph.D. in Computer Science with a dissertation on robust domain-invariant inertial localization in real-time federated edge computing. GPA 3.73/4.3, advised by Prof. Dongsoo Han.
```

### University of Engineering & Technology (UET), Taxila

B.Sc., Electrical Engineering. 2010 - 2014

```text
B.Sc. in Electrical Engineering with a thesis on real-time object detection and autonomous control using a 3-DoF robotic arm.
```

## Tech stack

- Kubernetes
- Azure Kubernetes Service
- GitHub Actions
- Python
- C++
- C
- PyTorch
- MLflow
- DeepSpeed
- Kafka
- FastAPI
- Docker
- CUDA
- vLLM
- NVIDIA Triton Inference Server
- ONNX Runtime
- SGLang

To contact Omer, message them from this profile with a Himalayas employer account: [Hire on Himalayas](https://himalayas.app/hire).

## People also viewed

- [Matheus Miranda](https://himalayas.app/@matheusmiranda): AI Engineer (Acting AI Lead), Tridge
- [sriman mohapatra](https://himalayas.app/@srimanmohapatra): AI Researcher (Graduate Researcher), Jeonbuk National University
- [andy l](https://himalayas.app/@andyl): MLsys engineer, Maum AI
- [Farrukh Olimov](https://himalayas.app/@farrukholimov): Machine Learning Engineer, Confidential
- [HS L](https://himalayas.app/@hsl): Student Intern, ECPlaza
- [Namgyu Youn](https://himalayas.app/@namgyuyoun): vLLM Contributor, vLLM
- [Choi Harim](https://himalayas.app/@choiharim): Researcher, Sejong Bid Institute
- [Usama Muhammad](https://himalayas.app/@usamamuhammad): AI Engineer, Agent Astro
- [jihun rho](https://himalayas.app/@jihunrho): AI Orchestration Program, SK Networks Family × Encore AI Campus
- [joon User](https://himalayas.app/@joonuser): AI Research Engineer, Tether
