Job Description

Job Description: Senior Scientist - Machine Learning (Audio)

Abu Dhabi

About Presight

Presight is an ADX-listed public company with Abu Dhabi based G42 as its majority shareholder and is the region's leading big data analytics company powered by GenAI. It combines big data, analytics, and AI expertise to serve every sector, of every scale, to create business and positive societal impact. Presight excels at all-source data interpretation to support insight-driven decision-making that shapes policy and creates safer, healthier, happier, and more sustainable societies. Today, through its range of GenAI-driven products and solutions, Presight is bringing Applied AI to the private and public sector, enabling them to realize their AI strategy and ambitions faster.

Position Overview

We're looking for an experienced and hands-on SeniorMachine Learning Scientist with deep expertise in audio and speech technologies to drive the design, development, and deployment of advanced ML models powering our data integration platform.

In this role, you will focuse on speech-to-text (with domain adaptation), speaker identification, keyword spotting, language identification, and deep fake detection. You will work closely with engineering and data infrastructure teams to build, optimize, and deploy scalable ML services in production using NVIDIA Triton, Kubernetes, and modern transformer-based architectures.

This position requires a unique combination of machine learning depth, software engineering proficiency, and team work skills to deliver robust, real-time, multimodal ML enrichment capabilities.

Key Responsibilities

Lead the ML Audio team in developing, training, and deploying production-grade models for:

Speech-to-text (including domain adaptation)
Speaker identification and verification
Keyword spotting
Language identification
Deepfake and synthetic audio detection
Research and integrate transformer-based and multi-modal model architectures into production pipelines.
Collaborate with platform engineers to deploy ML models via NVIDIA Triton Inference Server and ensure low-latency, scalable serving.
Design and maintain Airflow DAGs for data preprocessing, feature extraction, and model enrichment pipelines.
Ensure continuous improvement of models through retraining, performance monitoring, and data feedback loops.
Collaborate cross-functionally with Data Engineering, Backend, and Product teams to integrate ML capabilities into the platform.
Mentor and guide ML engineers and researchers, fostering a culture of technical rigor, creativity, and collaboration.
Oversee release and validation processes for ML components within larger system releases.
Contribute to and review production-level code in Python (C++, Golang, Rust knowledge is beneficial).

Requirements

Proven experience working within ML teams, ideally in audio/speech technology domains.
Strong foundation in machine learning and deep learning, particularly in speech processing and audio analysis.
Hands-on experience with:

Speech-to-text systems (ASR, domain adaptation, fine-tuning large models)
Speaker recognition and keyword spotting
Language ID and deepfake detection

Solid understanding of transformer-based model architectures (e.g., Wav2Vec2, Whisper, HuBERT, SpeechT5) and multimodal fusion techniques.
Proficiency in Python and familiarity with C++, Golang, Rust for performance-critical components is beneficial.
Experience deploying ML models in production using NVIDIA Triton and container orchestration platforms (e.g., Kubernetes).
Familiarity with Airflow for scheduling and orchestrating enrichment pipelines.
Understanding of data integration systems and ML-enriched workflows.
Excellent communication and leadership skills, with a focus on collaboration and cross-functional alignment.

Qualifications:

Bachelor's or Master's degree in Computer Science, Machine Learning, Electrical Engineering, or related field (PhD preferred).
Demonstrated record of leading ML initiatives from research to production.
Experience in startups or fast-paced, product-centric teams is highly desirable.

Experience:

5 – 7 years of combined software engineering and ML experience.
Proven success delivering speech or audio-based ML models in production environments.
Experience working with multi-language technology stacks and real-time inference pipelines.

Skills:

Speech-to-Text (ASR) & Domain Adaptation
Speaker & Keyword Recognition
Language Identification
Deepfake Detection
Transformer & Multimodal Architectures
NVIDIA Triton Inference Server
Airflow & ML Pipeline Orchestration
Kubernetes / Cloud Deployment
Programming: Python, C++
MLOps, CI/CD for ML Systems

Join us at Presight, where we offer a culture of innovation, outstanding career growth opportunities, and competitive rewards. If you're eager to conquer new frontiers in AI and thrive in a dynamic environment, we welcome you to our community.

What working at Presight offers:

Culture: An open, diverse and inclusive environment with a global vision that encourages personal growth and focuses on ground-breaking, industry-first innovations.

Career: Accelerate your career through high-impact projects and access to resources for continuous growth and learning opportunities.

Rewards: A competitive remuneration package with a host of perks including healthcare, education support, leave benefits and more.