Applied AI Engineer — Global Health & AI (REMOTE ROLE)

ICF•Rockville, MD
•$98,614 - $167,644•Remote

About The Position

ICF's Global Health and Development line of business is seeking an Applied AI Engineer to build frontier-model systems for global health and survey programs. The role develops AI-enabled tools for drafting survey reports, analyzing data through natural language, translating questionnaires into low-resource languages, and supporting interviewers and data quality in the field. Tools must match or exceed the expert-led processes they replace. The engineer will build the prompting, agent, and pipeline layer, along with evaluation harnesses that assess tool performance against ground truth created by experienced survey methodologists. The position is remote within the United States, with occasional travel, and is contingent upon award of program funding. The engineer will collaborate with technical and survey experts, connect models to data services and applications, monitor deployed systems, document results for public release, and help transfer tools to partner-country institutions.

Requirements

  • Bachelor's degree in Computer Science, Engineering, or a related field, or equivalent years of experience.
  • At least 4 years of professional software development experience, including building LLM-powered systems relied on by real users in production.
  • U.S. citizenship is required by the federal government contract.
  • Proficiency in Python or another modern programming language, such as JavaScript/TypeScript or Go, for back-end and AI application development.
  • Hands-on experience building with frontier-model APIs, including tool use and retrieval-augmented generation.
  • Demonstrated experience designing LLM evaluations, including test-set construction, automated and human grading, regression testing, and ship/no-ship decisions.
  • Working knowledge of LLM cost, latency, and model-selection trade-offs.
  • Experience building and deploying on a major cloud platform, such as AWS, Azure, or Google Cloud.
  • Demonstrated experience using agentic coding tools to specify, review, test, and take ownership of production code.
  • Experience with agent frameworks and AI evaluation tooling.

Nice To Haves

  • Experience with tasks where ground truth came from subject matter experts, such as statisticians, clinicians, scientists, or methodologists.
  • Experience with grounded generation over structured or tabular data, or LLM-assisted statistical analysis.
  • Experience with multilingual or low-resource-language NLP, translation, or speech.
  • Experience deploying open-weight models.
  • Experience with sensitive or regulated data, such as health data or human-subjects research.
  • Familiarity with model governance, monitoring, and responsible AI practice.
  • Experience in global health, international development, or low- and middle-income country settings.
  • Public work such as open-source contributions, published evaluations, or technical writing about systems built.
  • Ability to explain AI trade-offs clearly to technical and non-technical audiences, including funders, survey experts, and partner-government staff.
  • Ability to collaborate across disciplines, cultures, time zones, and organizational levels while treating domain experts as the source of truth.
  • Ability to communicate uncertainty, limitations, and negative results candidly.
  • Self-motivated and able to find ways to add value without much direct supervision.

Responsibilities

  • Design and build LLM pipelines and agents for survey workflows, including data-to-report, natural-language survey data analysis, multilingual questionnaire translation, and field-support tools for interviewers and supervisors.
  • Build evaluation harnesses as first-class deliverables; define acceptance criteria with the Technical Lead and survey experts, create test sets from expert-produced ground truth, and conduct non-inferiority comparisons to determine whether tools advance.
  • Route work across models based on reasoning complexity, volume, cost, latency, and data-residency requirements.
  • Manage inference costs by tracking token usage, context limits, prompt caching, and batch processing, and incorporate operating costs into model-selection decisions.
  • Connect models to data services and applications built by full stack developers.
  • Use agentic coding tools to produce software from clear specifications, then review, test, and take ownership of the resulting code.
  • Monitor production systems, conduct disciplined error analysis, and report positive and negative results.
  • Document methods and results for public release, including negative results.
  • Help transfer tools to partner-country institutions so they can operate them independently.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service