Artificial Intelligence and Machine Learning Scientist

Torch Technologies, Inc.•Colorado Springs, CO

1d•$130,000 - $138,000•Onsite

About The Position

Torch Technologies is seeking an Artificial Intelligence and Machine Learning Scientist to join a team researching, creating, and refining innovative approaches to the curation, labeling, typing, and fusion of Overhead Persistent Infrared (OPIR) data. The position supports the Advanced VizLab Division and involves collaboration with government, FFRDCs, and contractor partners to support advanced machine learning applications. The role includes designing and implementing an end-to-end out-of-distribution (OOD) detection and processing pipeline to triage, characterize, and extract value from large backlogs of unlabeled data; developing automated and human-on-the-loop labeling workflows to generate high-quality training datasets; and establishing data quality, provenance, and validation processes to improve model training, generalization, and operational performance. The scientist will also onboard lab users to the HEARTH system and support their model-development efforts. The successful candidate will contribute to a multidisciplinary team, build and integrate software components into functional systems, and deliver maintainable, testable, well-structured code. Responsibilities also include implementation of features and tools, documentation, support, automated testing, and low-level debugging. The position is onsite in Boulder, Colorado, follows an M–F, 8 a.m.–5 p.m. schedule, and requires 10–20% travel.

Requirements

U.S. citizenship.
A master’s degree in engineering with a concentration in artificial intelligence and machine learning or a related field, plus 8 or more years of applicable progressive experience; or a Ph.D. in engineering with a concentration in artificial intelligence and machine learning or a related field, plus 5 or more years of applicable progressive experience.
A Secret clearance at minimum to start, with the ability to maintain a Department of Defense clearance as a condition of ongoing employment.
Exceptional verbal, written, and interpersonal communication skills.
Ability to explain complex technical concepts and share knowledge with diverse internal and external audiences.
Effective problem-solving and debugging skills.
Ability to work independently, prioritize work, and meet deadlines.
Organization and ability to manage multiple projects simultaneously while understanding business needs.

Nice To Haves

Experience with software development and deployment in a Kubernetes cluster.
Passion for developing AI/ML applications.
Experience designing data contracts and integrating with APIs.
Experience developing CI/CD pipelines with Jenkins and GitLab.
Understanding of software code repositories such as Git and Perforce Helix.
Experience with issue tracking using Atlassian Jira.

Responsibilities

Collaborate with government, FFRDC, and contractor partners to research, develop, and refine methods for curating and labeling OPIR data.
Contribute to the typing and fusion of OPIR data for advanced machine learning applications.
Design and implement an end-to-end OOD detection and processing pipeline to triage, characterize, and extract value from unlabeled data backlogs.
Develop automated and human-on-the-loop labeling workflows to generate high-quality training datasets.
Establish data quality, provenance, and validation processes that improve model training, generalization, and operational performance.
Onboard lab users to the HEARTH system and support their efforts to develop models using it.
Contribute positively and effectively within the assigned team, lab, and company.
Deliver or contribute to a diverse range of features and systems.
Build and integrate software components into functional systems that meet or exceed customer expectations.
Develop maintainable, testable, well-structured, high-quality code while balancing fast iteration with long-term architecture.
Implement features and tools, provide documentation and support, write automated tests, and perform low-level debugging.