Backend Software Engineer; Product

Requirements

• Has 2+ years of experience building product-facing systems, with a strong sense of how backend decisions impact user experience
• Writes clean, maintainable code (Python preferred) and is comfortable working across a modern web stack (APIs, data systems, async workflows)
• Thinks in terms of product outcomes, not just technical implementation — cares about whether what they build is actually useful to users
• Is excited about working with AI/LLM-powered systems, and is interested in shaping their behavior in real‑world applications
• Takes ownership of problems end‑to‑end and collaborates well across product, design, and AI disciplines
• Is comfortable operating in ambiguous, fast‑moving environments and can iterate quickly on both product and technical direction
• (Desirable) Experience building or experimenting with LLM‑powered features (prompting, evaluation, RAG, etc.)
• (Desirable) Experience designing systems that balance latency, cost, and quality
• (Desirable) Exposure to user feedback loops, experimentation, or evaluation frameworks

• As a Software Engineer on our Product teams, you’ll work at the intersection of product, AI, and systems to shape how users interact with intelligence in real workflows
• You’ll work across the full lifecycle of AI‑driven features, from data ingestion and model orchestration to output structuring and delivery into user‑facing product surfaces
• This role requires balancing product thinking with systems design, ensuring that AI outputs are not only intelligent, but also reliable, performant, and aligned with real user needs
• You’ll partner closely with product, design, and AI teams to rapidly prototype, ship, and iterate on experiences like summaries, action items, and intelligent workflows and turning raw model output into high‑quality product features
• Build AI‑native product experiences end‑to‑end:
• Design and implement features that transform model output into structured, actionable user value—such as summaries, action items, and intelligent workflows
• Own the product quality loop:
• Continuously improve output quality by iterating on prompts, data flows, and backend logic, using real user feedback, evaluations, and experimentation
• Bridge product and AI systems:
• Translate product requirements into systems that orchestrate LLMs, retrieval, and structured data—shaping how intelligence behaves in production, not just integrating it
• Design for real‑world constraints:
• Make deliberate tradeoffs across latency, cost, reliability, and output quality to deliver fast, trustworthy experiences at scale
• Prototype and ship at high velocity using AI‑assisted development:
• Leverage state‑of‑the‑art tools (e.g., coding copilots, LLM‑driven workflows) to move quickly from idea to production while maintaining high engineering quality
• Build clean, scalable backend systems:
• Develop services in Python (e.g., Django) that support dynamic, AI‑driven product experiences and evolving system requirements