Senior AI Software Engineer

About the Role

We're building Direct differently than a traditional software company.

AI isn't something we're simply layering onto our engineering team to make developers a little faster. We're intentionally designing our engineering organization around AI performing a significant amount of development execution.

We're looking for a senior engineer who knows software deeply enough to make that model work.

You'll work closely with our CTO and Product team, leveraging advanced AI development capabilities to research, plan, build, test, and ship software at a pace that would normally require a much larger engineering team.

This isn't primarily a "write more code" role. We need someone with enough engineering depth to evaluate AI-generated work, recognize when it's taking the wrong approach, provide the right technical context and guardrails, and hold the output to senior engineering standards.

What You'll Own

AI Engineering Execution

• Leverage AI as a significant force multiplier for engineering work
• Give AI systems the context, constraints, and technical direction necessary to successfully execute
• Review AI-generated implementation plans, code, tests, and PRs
• Continuously improve how we use AI to increase development velocity without sacrificing quality

Architecture & Engineering Quality

• Protect architectural integrity as development velocity increases
• Identify poor patterns, unnecessary complexity, technical debt, and downstream risks before they enter production
• Establish and enforce engineering standards and reusable patterns
• Partner with the CTO on architecture, infrastructure, scalability, security, and major technical decisions

AI Accountability

• Never assume AI-generated output is correct simply because it works
• Challenge implementation decisions and understand why something was built the way it was
• Catch dependency, data, performance, security, and maintainability issues AI may miss
• Build automated checks and guardrails so quality becomes increasingly systemic rather than dependent on manual review

Product Delivery

• Work closely with Product to turn customer problems and acceptance criteria into shipped software
• Identify technical dependencies and risks before work begins
• Keep work moving from priority to production without unnecessary process or handoffs
• Help us operate with continuous delivery rather than traditional development bottlenecks

Continuous Improvement

• Build reusable knowledge, context, standards, and workflows that improve AI development performance over time
• Identify opportunities to automate work currently requiring humans
• Understand where AI development succeeds, where it fails, and continuously improve the system around it

What We're Looking For

You should be a genuinely strong software engineer first.

We're looking for someone who:

• Has senior-level experience building and maintaining production SaaS applications
• Understands architecture, APIs, databases, testing, cloud infrastructure, security, CI/CD, and modern development practices
• Can review unfamiliar code and quickly understand how systems fit together
• Has strong judgment around what "good" software looks like beyond whether the code simply works
• Is comfortable allowing AI to perform significant amounts of development execution
• Has meaningful hands-on experience with AI-assisted or agentic software development
• Naturally investigates dependencies and downstream consequences before approving changes
• Can move extremely quickly without becoming careless
• Doesn't need to personally write every line of code to feel ownership over the outcome

What This Role Is Not
This isn't a traditional developer role where tickets are assigned and you spend the sprint writing code.

It's also not a prompt engineering role.

And it's not someone supervising AI from the sidelines.

We want someone who understands software deeply enough to use AI as an incredibly powerful engineering resource while providing the technical leadership, context, standards, and accountability required to produce exceptional software.

How We'll Measure Success

You'll be successful when:

• We ship high-quality product improvements significantly faster than a traditional engineering organization
• AI can successfully execute increasingly complex work because our context, standards, and guardrails continue improving
• Development velocity increases without creating PR, QA, or technical debt chaos
• Production issues decrease as development velocity increases
• Product can move quickly without compromising architecture
• Our engineering organization scales through leverage rather than proportionate increases in headcount

Why This Is Interesting

We aren't trying to figure out how AI can help developers code a little faster.

We're asking a much bigger question:

What should a software engineering organization look like when AI can perform a significant amount of the actual execution?

You'll help us build that model.

We're looking for a small group of exceptional people who can use AI to produce outcomes that previously required a much larger engineering team.