Principal Software Engineer, AI Systems

ReClick.io offers a competitive compensation package for an on-site role in Brooklyn, NY, with a yearly salary range of USD 150,000 to 210,000. This US W-2 position places you at the heart of an AI powered SaaS platform for experiential marketing, where you will own end-to-end production AI pipelines, contribute to full-stack TypeScript development, and steer architecture decisions that shape the product.

In this senior role, you will collaborate in a culture that values transparency, strong technical opinions, and a pragmatic approach to building great customer experiences. You’ll be the principal technical voice on the team, influencing both the product direction and the quality of the implementation, while working hands-on to deliver fast, reliable software.

Responsibilities

• Lead delivery of a modern SaaS web application with AI integrations
• Architect features deployed to serverless and edge cloud functions, prioritizing correctness, performance, and flexibility for long running tasks
• Model complex problems in a relational database, manage schema migrations, optimize queries, and enforce data security with Row-Level Security
• Write production-grade TypeScript across the stack: Next.js App Router with React and TailwindCSS on the frontend, Supabase Edge Functions (Deno) on the backend, and PostgreSQL with RLS for data integrity
• Build and maintain the AI pipeline, an extensible orchestration system using Claude Haiku and Claude Sonnet that outputs structured results, with prompt caching and Zod validated schemas
• Implement and refine UI features using the Glow UI design system, ensuring fast and deliberate user interactions
• Manage CI/CD pipelines with Github Actions and Vercel webhooks
• Engage in product and UX discussions with a commitment to transparency and thoughtful rationale for decisions
• Develop test suites that provide meaningful feedback in a fast moving startup environment to prevent regressions

Requirements

• Have shipped at least one production system that calls an LLM with structured output, including prompt caching, Zod validated results, evaluation harnesses, and the ability to diagnose why a model produced an unexpected shape weeks after deployment
• Delivered full-stack TypeScript products in production
• Experience debugging a broken RLS policy
• Know when to use a database function versus application logic and when to make things configurable versus static
• Hold informed opinions about data fetching patterns and understand why optimistic updates matter for UX
• 6+ years building production web applications, with at least 1 year working on production AI/LLM systems
• Comfortable being the most senior technical voice in the room, ready to challenge ideas while executing decisions even if they add backend complexity for a better user experience

Technology

• TypeScript
• Next.js App Router
• React
• TailwindCSS
• Supabase Edge Functions
• Deno
• PostgreSQL with Row-Level Security
• RLS
• Zod
• Anthropic Claude Haiku
• Anthropic Claude Sonnet
• Glow UI
• SWR
• SWR Mutation
• Vitest
• Github Actions
• Vercel

Culture

• We maintain strong architectural opinions to move quickly without reinventing the wheel
• We move fast without cutting corners; clean builds, lint, and tests are essential
• We address problems openly and promptly when they could impact users at scale
• AI is a collaborator in our workflow, and you should be comfortable prompting, evaluating, and iterating on AI-generated code

Logistics

• W-2 employment in the US
• NYC-based, in-office in Brooklyn

Application Questions

1. Describe the most complex production system you've designed and shipped end-to-end. What was the hardest architectural decision you made, and what would you do differently now?
2. A link to something you've built (GitHub repo, deployed product, technical writeup, or portfolio piece). Add one sentence on what we should look at and why.