The opportunity
Replit is the agentic software creation platform that enables anyone to build applications using natural language. With millions of users worldwide, Replit is democratizing software development by removing traditional barriers to application creation.
What you'll do
Design, architect, and ship custom AI-powered internal platforms that serve: specific operational needs across Replit’s functional departments, including but not limited to People Operations, Customer Support, Sales, Recruiting, Marketing, Finance, and Executive Operations.
Translate ambiguous, often non-technical business requirements into technical: specifications, system architectures, and shipped software products.
Own platform development end-to-end, including front-end interfaces, back-end: services, data models, third-party integrations, and AI model orchestration.
Architect multi-stage AI pipelines incorporating large language models,: including prompt engineering, tool-call architecture, structured output generation, retrieval-augmented generation, and agentic workflows.
Conduct model selection across multiple frontier and open-source providers,: evaluating models against cost per query, latency, output quality, reliability, and task-specific performance.
Design and implement evaluation frameworks for AI features, including test: datasets, scoring rubrics, and regression tests, so that AI-driven workflows stay reliable in production.
What they're looking for
- You understand how a business actually works and can explain why each: department matters to the company winning.
- You have either run your own business or worked on a fast-scaling startup,: and you carry the ownership instinct that comes with it.
- You are comfortable with long hours and a very high pace, and you treat that: as part of the fun rather than a cost.
- You can engage credibly with both deeply technical people and non-technical: stakeholders, often in the same afternoon.
- You bring diplomacy, creativity, technical knowledge, and business acumen,: and you know when each one is the tool the moment calls for.
- You would rather ship something real and iterate than wait for perfect information.
- Prior experience as a founder, co-founder, or early team member at a: venture-backed startup, particularly one with an AI-native product.
- Prior experience in technology investment analysis or operational consulting,: with a demonstrated ability to evaluate business operations and identify high-leverage interventions.
