Principal Software Engineer, AI & Data Platform

About Elemynt

Elemynt develops secure AI infrastructure tailored for scientific and engineering research and development teams. Their platform integrates data, models, computational power, and expert workflows within environments emphasizing reliability, traceability, and stringent data control.

With a compact yet ambitious engineering team spread across Singapore and the United States, Elemynt’s mission is to transform advanced scientific computing into production-ready software tailored for practical use by technical teams.

Role Overview

This principal engineering role revolves around architecting and owning the data and AI engineering foundations that empower Elemynt’s platform. The focus is on transforming scientific and engineering datasets into reusable assets that serve analysis, model training, and automated workflows.

The position demands a technical leader who not only sets architectural direction but also actively codes complex components.

Responsibilities

• Design and build foundational data architectures for large-scale scientific and engineering results, ensuring clarity, queryability, reusability, and readiness for machine learning training.
• Develop domain-specific scientific data models supporting interactive exploration, automation, and machine learning applications.
• Engineer scalable data processing across diverse storage systems including object stores and analytical databases optimized for training.
• Implement data pipelines for tasks such as data curation, deduplication, formatting, creation of evaluation sets, and regression tracking.
• Develop, maintain, and operate training and fine-tuning pipelines for AI models applied to scientific and workflow-oriented products.
• Create intelligent workflow interfaces that seamlessly link user intentions, platform capabilities, and executable workflows while concealing system complexities from the end user.
• Lead model evaluation strategies, benchmarking, automated scoring, and quality tracking to enable measurable iterative improvements.
• Establish and enforce team-wide data and AI engineering standards, converting them into reusable codebases, documentation, and patterns.

Required Qualifications and Skills

• Bachelor’s or Master’s degree in Computer Science or closely related engineering fields.
• Over a decade of experience developing and delivering production-grade software systems.
• Proficiency in Python programming with proven ability to deliver full-system implementations.
• Extensive experience with large-scale data technologies including object storage, advanced analytical processing, training-optimized data formats, and robust production data pipelines.
• Hands-on in building data pipelines supporting model training, fine-tuning, evaluation, and ongoing enhancements.
• Direct involvement in training or fine-tuning machine learning models for structured outputs, workflow automation, or domain-specific functionalities.
• Strong grasp of relational, document, and columnar data models paired with informed decision-making on their application contexts.
• Comfortable operating within cloud, enterprise, and technical compute settings, familiar with distributed training or large-scale batch processing.
• Demonstrated ability to define and implement technical strategies in dynamic, early-stage environments.

Preferred (Nice to Have)

• Experience applying machine learning techniques to scientific datasets, including property prediction, generative modeling, graph-based methods, or simulation-related data.
• Knowledge of atomistic, materials science, chemistry, or engineering data ecosystems.
• Expertise with retrieval techniques for structured datasets, knowledge graphs, or hybrid search architectures.
• Background in crafting APIs or tool interfaces for reliable integration with intelligent systems.
• Experience developing complex data and machine learning workflows on production-grade orchestrators.
• Contributions to open source projects related to machine learning, data infrastructure, or scientific computing.

Location & Work Model

Based either in Singapore or the United States, with work being conducted either on-site or via a hybrid model depending on the specific location.

Additional Notes

Candidates who identify with the nature of this work are encouraged to apply even if not all criteria are met fully.