Fieldwire is seeking a Mid-Level AI & Machine Learning Engineer to help adapt and deploy AI models for construction document and image understanding. In this hybrid role based in San Francisco, you will build AI foundations that run in Fieldwire infrastructure and turn complex visual inputs into useful, context-aware information inside the product.
What you’ll do
- Architect, build, and maintain AI foundations across the full stack, embedding them into core platform capabilities
- Evaluate and select open-source models, then adapt them to domain-specific use cases using fine-tuning, transfer learning, and lightweight training
- Deploy and operate models within Fieldwire infrastructure, including GPU-backed environments with an emphasis on inference performance, resource utilization, throughput, and latency
- Develop solutions for image understanding including visual object detection, classification, and segmentation
- Extract and interpret information from PDFs, scanned documents, images, and other complex visual content by combining OCR, document layout analysis, and multimodal models
- Apply AI and computer vision techniques to surface relevant, context-aware information within users’ workflows
- Debug production issues across services and multiple layers of the stack
- Improve engineering standards, tooling, and processes
Required qualifications
- Bachelor’s or Master’s degree in Computer Science (or equivalent work experience)
- 3+ years of experience in a Fullstack, Backend, AI Engineering, or Machine Learning role
- Strong practical experience in machine learning, including training, evaluating, and deploying models
- Hands-on experience with computer vision and image-processing techniques
- Knowledge of modern toolchains used to build high-quality applications
- Expertise writing clean, well-tested, modular code
Technologies you may work with
- Open-source models
- Fine-tuning and transfer learning
- OCR and document layout analysis
- Multimodal models, transformers, and vision-language models
- GPU-backed environments
- Rust, Ruby on Rails, and Angular
Preferred interest areas
- Experience adapting pretrained models to domain-specific tasks through fine-tuning and transfer learning
- Understanding of modern model architectures, including transformers and vision-language models
- Experience with image and document processing, including OCR, document layout analysis, or extracting structured information from PDFs and images
- Experience building production systems, preferably in Rust, Ruby on Rails, or Angular
Location and compensation
This role is hybrid in San Francisco, California. Fieldwire’s headquarters are based in San Francisco, with an office centrally located right off of the Embarcadero BART station.
The estimated salary range is USD 170,000 - 195,000 per year, with offers determined based on skills, qualifications, and experience. The salary range represents the low and high end for the US market, and minimums and maximums may vary by location.
Benefits
- Corporate bonus ranging up to 30%
Fieldwire will consider this role based on the factors above, including your skills, qualifications, and experience.
