Director, Enterprise AI Business Applications

Hanwha Defense USA Inc•Arlington, VA

6h•$160,000 - $210,000•Onsite

About The Position

Hanwha Defense USA (HDUSA) develops and supplies advanced ground and sea combat systems to strengthen the capabilities of U.S. and allied forces. HDUSA has built an internal enterprise application ecosystem where AI is not a feature added at the end. Nearly every application we run was either built with AI or has AI integrated into how it works. We are looking for a Director of Enterprise AI Business Applications to own that ecosystem and scale it. This person leads the internal application team, serves as product owner for the applications we build ourselves, owns the SaaS and on-premise applications we buy, and is the go-to when anything in that portfolio breaks. This is a build role. The person in this seat ships software, reviews what the team ships, and sets the technical standard the rest of the team will work to. You are expected to build, and you are expected to build with AI. You must be able to architect a system, judge whether what came out of the model is any good, debug it when it is not, and stand behind it in production. That takes real software fundamentals, which is why we want someone with a genuine engineering and delivery background rather than a prompt operator. You will also be teaching. The people hired into this team after you will likely be earlier in their careers and faster with AI tooling than they are with software discipline. Setting that standard, and mentoring toward it, is a core part of the job. We are looking for someone who has stayed with their systems long after launch, through upgrades, breaking dependency changes, data migrations, security findings, and outages. Getting a prototype working is the easy part. We need applications that are still running and still maintainable in three years.

Requirements

Software engineering background with real depth. You have written production code, owned systems in production over multiple years and at meaningful user scale, and been responsible when they failed.
8 or more years of professional software engineering experience, including at least 3 years accountable for the output of other engineers.
You have inherited systems you did not write and had to maintain, refactor, or replace them.
You have set engineering standards that others followed, such as code review, testing, and release practice, and held a team to them when it was inconvenient.
Experience leading software delivery, whether as an engineering manager, tech lead, or founding engineer. You have been accountable for what a team shipped.
Hands on CI/CD experience. You have built pipelines, not just used them.
Strong understanding of agentic AI systems. You know how current frameworks work, how agents are orchestrated, how agent to agent collaboration is structured, what the integration strategies are, and where each one breaks down.
You have built and shipped applications using modern AI, and you use AI as a working tool rather than as a subject you follow.
Product ownership experience. You can take an ambiguous business problem, define what should be built, and drive it to something people actually use.
Experience owning third party applications, including selection, integration, vendor management, and support.
Security minded by default in how you design applications and AI systems.
Comfortable initiating conversations with business leaders, and skilled at pulling ideas and feedback out of them that convert into valuable products and features.
Willingness to be hands on keyboard. This role builds alongside the team.
U.S. citizen and able to obtain a U.S. Government security clearance.

Nice To Haves

Experience in a regulated industry such as defense, aerospace, financial services, healthcare, or critical infrastructure.
Experience building on enterprise data, including integrating applications with ERP, data warehouses, and internal APIs.
Experience standing up an application or platform team from a very small starting point.
Familiarity with government cloud environments and the constraints that come with them.
Manufacturing, engineering, or program execution domain knowledge.
Experience mentoring junior engineers, particularly engineers who came up on AI assisted development.

Responsibilities

Own the full portfolio of HDUSA business applications, both internally built and externally sourced, across their lifecycle from selection or design through sustainment and retirement.
Serve as the escalation point for any application issue in that portfolio. When something is broken, this role owns getting to resolution regardless of whether we wrote the code or bought it.
Evaluate, select, integrate, and manage point solutions, from single purpose SaaS tools to major platforms. Own the vendor relationship, the integration, the license position, and the decision to keep or replace.
Maintain a clear view of what the company runs, what it costs, who owns it, what data it holds, and what it connects to.
Financial understanding of product ROI and being able to deploy that across the ecosystem to ensure tangible value is being created
Act as product owner for internally built enterprise applications. Set the roadmap, write and prioritize the backlog, define requirements, and make the call on scope while managing scope creep in an aggressive agentic AI development environment
Work directly with business stakeholders across Finance, Contracts, Supply Chain, Manufacturing, Engineering, Program Management, Security, IT and HR to understand how work actually happens before designing a system around it.
Own adoption, not just delivery. Shipped applications that go unused are non-value add.
Define success measures for each application and report on them.
Own the software delivery lifecycle for the internal team, including source control strategy, branching, code review, environment management, testing, release, and rollback.
Build and maintain CI and CD pipelines. Automated build, test, and deployment is expected as the baseline, not an aspiration.
Establish standards for logging, monitoring, error handling, and observability so that problems surface before users report them.
Ship code personally. Review the team code. Keep technical debt visible and managed rather than accumulating quietly.
Own production operations for the internal portfolio, including incident response, root cause analysis, uptime, backup and restore, and recovery testing.
Own long-term maintainability, including dependency and framework upgrades, deprecations, data migrations, and backwards compatibility. What this team builds has to survive its own success.
Hold documentation and handover quality high enough that systems outlive the people who built them.
Own the architecture for how AI is built into HDUSA applications, including model selection and routing, retrieval over internal data, tool and API calling, context management, and evaluation.
Design and implement agentic systems, including multi agent patterns, agent to agent communication and collaboration, orchestration across services, and the handoff and escalation logic between automated steps and human decision points.
Choose and apply the right frameworks and integration strategies for the problem, and know the tradeoffs well enough to defend the choice.
Build evaluation and testing into AI features. Know how a given application performs, where it fails, and what happens when it does.
Keep security in the design from the start, including prompt injection and tool abuse exposure, agent permission scoping, data leakage between contexts, third party model and vendor risk, and audit trails for automated action.
Ensure applications and AI capability meet CUI, ITAR, export control, and FOCI mitigated access requirements, including how data is classified and where it is allowed to be processed.
Work within HDUSA AI governance, including routing of sensitive and export-controlled data to approved environments and control over model and tool access.
Partner with IT Security and Compliance on application security review, access control, logging, and evidence for audit.
Apply the same standard to purchased software as to what we build, including data residency, retention, subprocessors, and where the vendor sends our data.
Lead and grow the enterprise application team. Hire, onboard, set standards, and mentor.
Scale the application ecosystem aggressively, from its current footprint toward an enterprise wide and eventually global capability, without letting quality, security, or maintainability degrade as volume increases.
Build the ecosystem so that applications share patterns, services, identity, and data rather than becoming a collection of disconnected tools.
Provide clear status, risk, and roadmap reporting to senior leadership.
Other duties as assigned.