Applied Data Scientist & ML Platform Engineer

Drill Insight is a premier provider of AI-powered technology services and digital subsurface solutions headquartered in Perth, Australia, with active operations and offices across the North Sea, UK, Middle East, Asia Pacific, and Australia — including strategic hubs in Dubai, Kuala Lumpur, and India.

At Drill Insight, we deliver Data-Driven Insights – Powered by AI. We specialise in developing hybrid machine learning (ML) models, generating subsurface intelligence, enhancing reservoir characterisation, optimising field development and drilling, and ultimately maximising production outcomes for our global clients.

Having successfully executed over US $22.6 million in contracts, our impact is driven by an elite team of 16 multidisciplinary experts spanning data science, artificial intelligence, geoscience, well engineering, reservoir engineering, geology, geophysics, and project management. Our strength lies in integrating traditional oilfield expertise with cutting-edge ML/AI and digital technologies, enabling unparalleled optimisation across subsurface, drilling, and production domains.

Our AI/ML-driven Field-Insights, Well-Insights, and Drill-Insights platforms support Oil and Gas E&P companies’ planning, operational and digital objectives. Our integrated team of drilling engineers, geoscientists, and data scientists has delivered AI/ML solutions for Chevron, Woodside, Equinor, Aker BP, SNOC and others - helping operators forecast drilling hazards, extract insight from offset wells, refine drilling targets, trajectories, well placements, and improve well performance and deliverability across varied basin conditions. Our integrated solutions, Field-Insights, Well-Insights, and Drill-Insights, can deliver measurable value through Field Development Planning, Cloud-Based Well Review Automation, Real-Time Drilling Visualisation, Performance Evaluation, Benchmarking, and an Optimisation Platform.

We are looking for builders rather than title collectors—people who can work across disciplines, challenge assumptions, communicate clearly with drilling engineers and turn difficult operational problems into reliable production systems.

Oil and gas experience is valuable, but exceptional candidates from industrial AI, robotics, mining, autonomous systems, process control, aerospace or real-time optimisation are strongly encouraged to apply.

This is an opportunity to:

• develop genuinely new industrial AI capability;
• work directly with experienced drilling, directional and subsurface specialists;
• contribute to potential intellectual property and future technical publications;
• build systems that could materially improve safety, drilling efficiency and well delivery.

Please send your CV or LinkedIn profile, availability, and a brief description of the hardest technical system or model you have built to: ***email_hidden***

Subject: AI Drilling Programme

A GitHub profile, technical paper, portfolio, architecture diagram or short project summary is welcome—but not mandatory.

The location and engagement model will be discussed with the shortlisted candidates.

Role Description: Drill Insight is seeking an Applied Data Scientist & ML Platform Engineer to lead the modelling workstream for an intelligent drilling optimisation programme, available on either a hybrid or fully remote basis. The role may be based in Perth, Western Australia, or performed remotely from Thailand, Vietnam, the Philippines, Kuala Lumpur, or India with occasional travel to Australia or client. Candidates based in Australia may also be offered flexible working arrangements, including the option to work from home for part of the week.

The opportunity

This is a multidisciplinary role for someone who enjoys working across data science, data engineering, feature engineering and lightweight MLOps. The successful candidate will transform heterogeneous drilling, directional, BHA, bit, mud and subsurface information into trusted datasets and reusable ML pipelines. This is not a reporting-only analyst role.

Key responsibilities

Data readiness and engineering

· Inventory and assess historical and real-time project datasets.

· Ingest and process WITS/WITSML, EDR, DDR, MWD/LWD, survey, BHA, bit and subsurface data.

· Standardise units, depth references, timestamps and channel naming.

· Align time, measured depth, formation, operation state, BHA and bit-run information.

· Handle missing values, duplicates, outliers, sensor spikes and sampling-rate differences.

· Build a lightweight project data lake, metadata register and data dictionary.

· Produce repeatable data-quality and readiness reports.

Feature engineering and modelling

o MSE,

o ECD,

o SPP and pressure deviations,

o torque-and-drag residuals,

o vibration indices,

o ROP efficiency,

o formation and BHA context,

o directional changes,

o lagged and rolling-window variables.

• Build training, validation and blind-test datasets.
• Implement and tune models under the direction of the Principal AI/ML Lead.
• Produce feature-importance and explainability outputs.

ML platform and lightweight MLOps

• Maintain experiment tracking and dataset versioning.
• Build reusable training and evaluation pipelines.
• Manage the model registry.
• Develop controlled retraining workflows.
• Monitor model performance and data drift.
• Support Docker packaging and production deployment.

Essential capabilities

• Strong Python, pandas/Polars and SQL.
• Hands-on experience preparing complex real-world datasets.
• Experience building repeatable ML pipelines.
• Understanding of supervised learning, anomaly detection and time-series features.
• Familiarity with MLflow or an equivalent experiment/model-management platform.
• Experience with Git, Docker and reproducible environments.
• Ability to work independently with incomplete and imperfect operational data.

Qualifications

• Bachelor's degree or equivalent experience in a quantitative field (Statistics, Mathematics, Computer Science, Engineering, etc.)
• At least 1 - 2 years of experience in quantitative analytics or data modelling
• Deep understanding of predictive modelling, machine learning, clustering and classification techniques, and algorithms
• Fluency in a programming language (Python, C,C++, Java, SQL)
• Familiarity with Big Data frameworks and visualisation tools (Cassandra, Hadoop, Spark, Tableau)

What success looks like

• Traceable, model-ready datasets.
• Reliable time-depth-BHA-formation alignment.
• Automated QC and feature pipelines.
• Reproducible models that can be retrained as new wells are drilled.