AI Scientist / Machine Learning Engineer

Full Description

Responsibilities

Train and fine‑tune AI models (vision and language) for document layout understanding, OCR enhancement, visual data extraction, and captioning / alt‑text generation.

Optimize AI pipelines for real‑time or near‑real‑time performance, replacing high‑latency models with efficient alternatives.

Develop and integrate models trained on real‑world and synthetic datasets.

Create or annotate datasets for graph / table classification, layout segmentation, and text extraction.

Evaluate models for accuracy, generalization, and inference speed.

Collaborate with software engineers to deploy models in production environments.

Participate in the development of AI‑assisted user interfaces for manual correction, QA, and accessibility validation.

Maintain research rigor through documentation, testing, and performance benchmarks.

Requirements

PhD / MSc in Computer Science, Artificial Intelligence, or related field.

Strong research and hands‑on experience in computer vision (OpenCV, object detection, layout parsing), NLP (captioning, text generation, alt‑text generation), and deep learning (PyTorch, TensorFlow).

Strong programming skills in Python, with experience building end‑to‑end AI pipelines (data ingestion, preprocessing, training, inference, deployment).

Model optimization and performance tuning (e.g., quantization, pruning, batch inference).

Training and evaluating models on noisy, scanned, or complex real‑world documents.

Dataset creation / augmentation for domain‑specific tasks.

Familiarity with or willingness to learn tools like LayoutParser, LayoutLMv3, BLIP / LLaVA, YOLO, Tesseract.

Experience with lightweight front‑end frameworks for ML interaction (e.g., Streamlit).

Bonus Skills

Knowledge of accessibility standards (e.g., WCAG, EN 301‑549).

Experience deploying AI in secure, high‑throughput production environments.

Background in OCR QA or document remediation tools.

Working knowledge of containerized API deployment (Docker, FastAPI, Hugging Face Transformers).

To Apply

A brief cover letter explaining your relevant experience in the domains mentioned above.

(Optional) Links to publications, GitHub, or portfolio projects.

#J-18808-Ljbffr