Artificial Intelligence Specialist

Stelvio Inc.; Austin (TX); On-site; USD 180,000 - 260,000; Full time

Job summary

Austin-based Stelvio Inc. is seeking Principal/Staff-level AI/ML Engineers to take significant ownership of core model-development work. This highly autonomous role focuses on building, training and optimizing AI models rather than simple integration.

You will work closely with founders and the leadership team to design experiments, move models from early development through production deployment and shape the future of the company’s AI capabilities.

Qualifications

• Five+ years in AI/ML, research, or related software engineering.
• Built, trained or materially modified machine-learning models.
• Experience deploying models in production environments.
• Advanced Python development skills.
• Commercial experience with PyTorch, JAX, TensorFlow or comparable frameworks.
• Strong understanding of modern deep-learning and model-training techniques.
• Experience measuring and improving model performance.
• Ability to work independently within technically ambiguous environments.
• Experience delivering AI capabilities within a commercial or B2B setting.

Responsibilities

• Designing and training specialist machine-learning models.
• Building models from the ground up where the problem requires it.
• Fine-tuning and post-training foundation models using proprietary datasets.
• Owning experimentation, evaluation, benchmarking and model validation.
• Taking research concepts through to reliable production systems.
• Improving accuracy, latency, throughput, reproducibility and infrastructure efficiency.
• Optimizing model training and inference across GPU environments.
• Contributing to proprietary and potentially patentable AI technology.
• Helping define future model architecture and technical direction.
• Maintaining direct ownership of your work from initial concept through deployment.

Skills

Python

PyTorch

JAX

TensorFlow

Deep learning

Independent work

Production deployment

Model evaluation

Commercial AI

Tools

CUDA

Job description

Austin, Texas | Full-Time | On-Site - moving to Hybrid

The strongest AI engineers are rarely looking for another job. They are usually looking for a technical problem worth owning.

Stelvio is supporting an Austin-based company seeking Principal/Staff-level AI/ML Engineers to take significant ownership of its core model-development work.

This is a highly autonomous individual-contributor position for someone who wants to build - not simply integrate - AI technology.

Working closely with the founders and technical leadership team, you will be trusted to identify the right technical approach, design experiments and take models from early development through optimization and production deployment.

You will be responsible for areas including:

• Designing and training specialist machine-learning models
• Building models from the ground up where the problem requires it
• Fine-tuning and post-training foundation models using proprietary datasets
• Owning experimentation, evaluation, benchmarking and model validation
• Taking research concepts through to reliable production systems
• Improving accuracy, latency, throughput, reproducibility and infrastructure efficiency
• Optimizing model training and inference across GPU environments
• Contributing to proprietary and potentially patentable AI technology
• Helping define future model architecture and technical direction
• Maintaining direct ownership of your work from initial concept through deployment

This is not a position where you will inherit narrowly defined tickets or operate several layers away from technical decision-making. You will have the freedom to challenge assumptions, explore new approaches and influence how the company builds its core AI capabilities.

What we’re looking for:

• At least five years of professional AI/ML, research or closely related software-engineering experience
• Strong evidence of personally building, training or materially modifying machine-learning models
• Experience fine-tuning and deploying models within production environments
• Advanced Python development skills
• Commercial experience with PyTorch, JAX, TensorFlow or comparable frameworks
• A strong understanding of modern deep-learning and model-training techniques
• Experience measuring and improving model performance
• The ability to work independently within technically ambiguous environments
• Experience delivering AI capabilities within a commercial or B2B setting

A key area of interest is practical experience improving the performance of computationally intensive models.

This could include:

• CUDA or GPU-accelerated model development
• Training and inference optimization
• Distributed or multi-GPU training
• Quantization, compilation or model compression
• Memory and compute optimization
• High-throughput or low-latency model serving
• Profiling and resolving model-performance bottlenecks