Director of Agentic Software Development

FactSet•Boston, MA

10d•$200,000 - $260,000

About The Position

FactSet is driving a transformation to agentic software development. The future is one where AI is not an add-on to the engineering workflow but the primary way software is built, tested, deployed, and maintained. This role leads that transformation end-to-end. The Director of Agentic Software Development owns the strategy, enablement, and measurement required to move FactSet's development lifecycle from AI-assisted coding to AI-native software delivery. This is not a tools procurement role. It is a leadership role that requires a clear point of view on what agentic development and the AI-SDLC should look like across the full lifecycle, and the ability to drive the organization toward it. The role is accountable for changing how FactSet builds software, with real metrics, real adoption, and real impact on delivery velocity and quality.

Requirements

Bachelor's degree or higher in Computer Science or related technical field
Technical fluency in modern software development and how AI tooling integrates across the lifecycle
Hands-on familiarity with the AI development tools landscape, including agentic coding frameworks, LLM-powered development, and the infrastructure that supports them
Track record delivering platform capabilities or transformation programs in an enterprise environment
Experience leading in fast-moving environments where the work itself is still being defined
10+ years of relevant experience

Nice To Haves

Deep understanding of the full software delivery lifecycle, not just engineering, and a clear vision for how AI reshapes it
Proven ability to drive adoption of new practices across a large organization, not by mandate but by building trust, evidence, and momentum
Comfort operating in ambiguity where the destination is clear but the path is being defined in real time
Natural connector across teams, levels, and functions, equally effective with senior engineers, product leaders, and executives

Responsibilities

Define and execute the roadmap for FactSet's transition to agentic development.
Own the vision for how AI reshapes every stage of the software delivery lifecycle, from requirements decomposition through code generation and review to testing, security, deployment, and operations.
Set the pace and sequence for the organization, balancing ambition with the trust-building required for teams to adopt fundamentally new workflows.
Shape the portfolio of AI development tools across FactSet and the emerging ecosystem of agentic coding platforms.
Bring the transformation perspective to build-vs-buy decisions.
Inform which tools are core to FactSet's developer platform, which are experimental, and which should be retired.
Stand up and lead the governance structure for agentic software development across FactSet.
Identify and leverage the right leaders to drive progress, maintain alignment, and ensure that what works in one team flows to the rest.
Connect the people already doing the work, give them a forum to share and decide together, and keep the transformation on track as it scales.
Build the infrastructure that makes agentic development effective at FactSet specifically. This includes MCP servers and skills that give AI tools access to FactSet's standards, libraries, and institutional knowledge; verification and trust layers that catch hallucination and enforce code quality; plugin marketplaces and skills registries that allow teams to share what they build; and learning curricula that bring every engineer along the adoption curve.
Build and sustain the internal community around agentic development.
Create forums for sharing wins, surfacing gaps, and spreading best practices across engineering organizations.
Work with early adopters and skeptics alike.
Own the strategy for how FactSet allocates, tracks, and governs token and compute spend as agentic software development workflows scale.
Ensure spend is visible by team and project, budgets are calibrated to adoption pace, and cost governance is built into the workflow rather than applied after the fact.
Define what success looks like and own the numbers. Metrics should capture what matters: engineering velocity (time from requirement to production), code quality, developer experience, and the share of engineering work that is AI-native vs. manual.
Carry real visibility and accountability for results.