Togetherwork

Software Engineer

• Togetherwork is seeking a Principal Architect, AI & Developer Productivity to own how AI accelerates the software development lifecycle across the portfolio. This is a hands on leadership role for someone who has shipped AI augmented engineering tooling at scale and can prove measurable improvements in developer throughput, software quality, and cycle time.
• You will define and operationalize the AI assisted development stack across the organization: IDE assistants, code review automation, test generation, security scanning, documentation, and release automation. You will set the standards, build the platform, and drive adoption across product teams. You will measure outcomes against DORA metrics and retire tools that do not produce returns, regardless of how fashionable they are.
• Define and operationalize the AI assisted engineering platform across the portfolio, covering IDE assistants, agentic coding tools (Claude code, cursor, etc), code review automation, test generation, security scanning, documentation, and release automation.
• Architect a model and vendor agnostic abstraction layer so the organization is not locked into a single tool, model, or provider as the landscape evolves monthly.
• Establish reference architectures and golden paths for AI augmented workflows that teams can adopt without forcing a single stack across all products.
• Establish acceptable use, IP protection, intellectual property leakage prevention, secret scanning, and data exfiltration controls for AI in the SDLC.
• Implement open source license scanning to prevent contamination from AI generated code that reproduces GPL, AGPL, or other restrictive license material.
• Define audit trail and traceability standards: which AI tool wrote what code, what tests were generated, what was reviewed, what was approved.
• Partner with Security, Legal, Compliance, and Risk to embed SOC 2, PCI, PII, SOX, data residency, and other regulatory requirements into the platform design.
• Support audit and risk assessment readiness by ensuring platform documentation, logs, and controls meet enterprise and regulatory expectations.
• Embed AI driven capabilities into CI/CD: automated pull request review, test synthesis, flaky test triage, vulnerability remediation, intelligent rollout, and incident analysis.
• Establish quality gates for AI generated code including coverage, mutation testing, security scanning, and license compliance before merge.
• Lead enablement across product teams: onboarding paths, paved roads, internal developer portal capabilities, and training for AI assisted workflows.
• Treat developer experience as a product with clear roadmaps, success metrics, user research, and feedback loops.
• Distinguish real productivity from the illusion of productivity. AI tools inflate volume metrics without necessarily delivering value, and traditional metrics like commits and lines of code are unreliable in AI native workflows.
• Report tool cost against measured outcomes. Make kill, scale, or replace decisions on tools that do not return $2 to $3 of value for every $1 of cost.
• Maintain an evaluation harness so new tools can be benchmarked against incumbents on real internal work, not vendor demos.
• Evaluate and select tooling across the current market: GitHub Copilot Enterprise, Cursor, Claude Code, and emerging entrants. Negotiate enterprise terms in partnership with procurement.
• Make defensible build vs buy decisions on AI components, frameworks, and pipeline integrations based on cost, security posture, switching cost, and outcomes.
• Stay current on emerging tools and models. Recommend platform evolution quarterly rather than annually. The field moves monthly.
• Bring acquired engineering teams onto the standard AI augmented SDLC platform with a clear runbook for tooling rationalization.
• Evaluate acquired company SDLC tooling and provide structured recommendations on what to integrate, rationalize, or retire.
• Own the total cost of AI in the SDLC: license consumption, token spend, infrastructure, and developer time. Implement chargeback, cost ceilings, observability, and alerting.
• Manage token spend at scale.
• Build cost models for new tool rollouts that include training, change management, and ongoing platform support, not just license fees.
• Partner with engineering leaders, product, security, legal, and procurement to align platform direction with business strategy.
• Mentor senior engineers and engineering managers on AI assisted development patterns and the discipline required to use them effectively.
• Communicate architecture decisions, trade offs, and platform outcomes clearly to executive stakeholders including the CTO and CFO.

Principal Architect, AI – Developer Productivity