Machine Learning Engineer

Amniscient LLC; United States; Híbrido; USD 120.000 - 180.000; Jornada completa

Descripción de la vacante

Amniscient LLC is seeking a Machine Learning Engineer II to design and scale CV systems that turn images into labeled datasets, train on GPU clusters, and ship models to production with a single click.

You will own features, collaborate with data/platform/product teams, and gain deep insight into computer vision models while growing alongside senior engineers.

Formación

• 2-5 years of professional experience building ML systems.
• At least one object detection model shipped to production.
• Strong Python skills and hands-on experience with PyTorch.
• Working knowledge of modern detection architectures (one-stage, two-stage, transformer-based).
• Solid grasp of detection fundamentals: IoU, mAP, NMS, and class imbalance.
• Experience with image augmentation, dataset curation, and annotation quality control.
• Comfort with Git, Linux, Docker, and training on GPUs.
• BS in Computer Science, Computer or Electrical Engineering, or related field.

Responsabilidades

• Train, fine-tune, and evaluate object detectors for production use cases.
• Build and maintain data pipelines for image and video datasets, including labeling workflows and quality checks.
• Define evaluation metrics (mAP, per-class precision/recall, IoU, latency) and perform error analysis.
• Optimize models for deployment through quantization, pruning, and export to ONNX or TensorRT.
• Deploy and monitor models in production, tracking accuracy drift and data distribution shift.
• Run experiments with clear tracking and reproducible results.
• Write clean, tested Python code and participate in code reviews.
• Share findings with product and engineering stakeholders in plain terms.

Conocimientos

Python

PyTorch

Git

Linux

Docker

Object detection

Data pipelines

Image augmentation

Educación

BS in Computer Science / Computer or Electrical Engineering

Herramientas

OpenCV

Albumentations

Pillow

ONNX Runtime

TensorRT

Triton Inference Server

Descripción del empleo

Imagine a warehouse team, a farmer, or a city planner training a state-of-the-art object detector on their own images, without needing an ML team. That's what we're building, and we'd like you to help build it.

As a Machine Learning Engineer II on our ML team, you'll design the systems that turn raw images into labeled datasets, run training at scale on GPU clusters, benchmark every model, and put it into production with a single click. Every improvement you ship gives thousands of customers a better model.

You'll own real features from day one, work closely with data, platform, and product teams, and have senior engineers in your corner as you grow. Along the way, you'll gain deep insight into how computer vision models work and how to make them better.

WHAT YOU'LL DO

• Train, fine-tune, and evaluate object detection models (e.g., YOLO-family, DETR-family, Faster R-CNN) for production use cases
• Build and maintain data pipelines for image and video datasets, including labeling workflows and quality checks
• Define evaluation metrics (mAP, per-class precision/recall, IoU, latency) and run error analysis to decide what to fix next
• Optimize models for deployment through quantization, pruning, and export to ONNX or TensorRT
• Deploy and monitor models in production, tracking accuracy drift and data distribution shift
• Run experiments with clear tracking and reproducible results
• Write clean, tested Python code and take part in code reviews
• Share findings with product and engineering stakeholders in plain terms

WHAT YOU'LL BRING

• 2-5 years of professional experience building ML systems, with 2+ years focused on computer vision
• At least one object detection model shipped to production and maintained after launch
• Strong Python skills and hands-on experience with PyTorch
• Working knowledge of modern detection architectures, including one-stage, two-stage, and transformer-based detectors
• Solid grasp of detection fundamentals: anchor-based and anchor-free methods, IoU, Non-Maximum Suppression, mAP, and class imbalance
• Experience with image augmentation, dataset curation, and annotation quality control
• Comfort with Git, Linux, Docker, and training on GPUs
• BS in Computer Science, Computer or Electrical Engineering, or a related field

NICE TO HAVE

• MS or PhD with a computer vision focus
• Experience with multi-object tracking, instance segmentation, or video-based detection
• MLOps experience: experiment tracking, model registries, CI/CD for models, and production monitoring
• Distributed training across multiple GPUs
• Edge deployment (NVIDIA Jetson, mobile, or embedded) and inference optimization with ONNX Runtime or TensorRT
• Publications, open-source contributions, or competition results in computer vision

OUR TECH STACK

• Language: Python
• Image processing: OpenCV, Albumentations, Pillow
• Inference: ONNX Runtime, TensorRT, Triton Inference Server

THE DETAILS

• Team: Machine Learning, reporting to Justin Serowik
• Location: Houston, TX, remote, or hybrid
• Start date: As soon as possible