Job Overview
The Applied AI Engineer (AAE) ensures AI systems work in the real-world contexts they are designed to serve. This works on designing (e.g., building flows and structured agents), deploying, adapting, testing, and validating AI systems in program environments to ensure they are accurate, usable, and effective in practice.
The AAE works directly with country teams, partners, and communities to configure AI systems for local contexts, including language, cultural nuance, and operational constraints. They are responsible for what goes into the system, how it behaves, and whether it delivers meaningful outcomes for end users, making key implementation decisions on system configuration, behavior and deployment approaches based on field conditions and user needs.
Owning deployments across the full lifecycle, from initial configuration through iteration, evaluation and scale, the AAE translates real-world complexity and field insights into concrete system improvements, product decisions and deployment strategies.
Acting as the bridge between technical development and field implementation, the AAE surfaces failure modes early, strengthens system performance and informs the evolution of tools, workflows and infrastructure required for scalable deployment. The role also identifies patterns across implementations and translates them into reusable frameworks and best practices to improve speed, quality, and consistency across programs.
By enabling high-quality deployments at speed, this role ensures that IRC's investments in AI translate into tangible impact for the people we serve.
Major Responsibilities
AI Deployment, Configuration & Context Adaptation
Lead end-to-end deployment of AI systems in program contexts, from initial scoping and configuration through live use
Adapt systems to local environments, including language, cultural context, and operational realities
Configure prompts, workflows, and AI System behavior, including conversation flows, instructional logic and user experience Curate, structure, and validate domain-specific knowledge bases including system memory and personalization strategies to improve relevance and continuity
Ensure systems reflect real-world humanitarian knowledge that may not exist in training data
Support multiple concurrent software deployments
Testing, Validation & Continuous Improvement
Test AI systems in real-world conditions to identify failure modes before scale
Applies an agile, iterative approach, rapidly building, testing, and refining systems based on real-world feedback, while exercising strong judgment on when to iterate vs. escalate or rethink approach
Conduct safety validation and red teaming to identify risks, harms, and unintended consequences ensuring responsible and ethical AI deployment in vulnerable contexts
Troubleshoot technical and operational issues, including in low-connectivity environments
Collect and interpret user and community feedback on usability, relevance, and performance
Escalate systemic issues and collaborate with engineering teams on fixes and improvements
Define and track success metrics to evaluate system performance, user engagement and real-world impact over time
Partnership & Cross-functional Collaboration
Train field teams on deploying systems and managing structured handovers to ensure adoption and sustained use
Document deployment processes, configurations, and lessons learned to build institutional knowledge and support replication
Work closely with AI engineers and technical teams to translate field insights into system improvements
Collaborate with program teams and partners to align deployments with priorities
Act as the primary interface between field teams and technical development during deployment
Support scoping and technical input for proposals and cost recovery opportunities where deployment capacity is a factor
Ensure solutions are integrated into workflows in ways that are practical, adopted, and sustained
Contribute to shared learning across deployments to improve future implementation
Identify patterns across deployments and translate them into reusable frameworks and best practices to improve speed and quality for future implementations
Contribute to product design and system improvements by translating field insights into system requirements, feature priorities and roadmap input
Key Working Relationships:
Position Reports to: [Sr. Deployment Engineer]
Key Relationships: AI/Technology development teams, IT, Data, regional and country program teams, implementing partners, sector specialists
Requirements:
3 - 5+ years of experience in software engineering, technical implementation, or related roles
Proficiency in JavaScript/TypeScript and modern web development practices; comfort with Python or similar scripting languages a plus.
Experience deploying and iterating on LLM-based applications in real-world environments
Experience with communications or messaging platforms (e.g., Zendesk, WhatsApp, Telegram) preferred
Experience integrating AI systems with third-party platforms and business tools (e.g. CRMs, messaging platforms, APIs), with familiarity with common integration patterns, authentication approaches, and data flow considerations.
Proficiency with AI-assisted development tools (e.g. Cursor, Claude Code) and a clear-eyed understanding of their limitations and failure modes.
Strong problem-solving skills, with the ability to translate real-world challenges into technical solutions
Experience working in low-resource, high-complexity, or field-based environments
Ability to operate in ambiguous environments and define structure where none exists
Strong understanding of how AI systems behave in practice, including limitations and failure modes
Ability to work across technical and non-technical teams and translate between them
Experience training non-technical users and managing system handovers
Experience in humanitarian, development, or nonprofit contexts strongly preferred
Familiarity with East or Central Africa contexts preferred
Fluency in English required; additional regional language skills preferred