What you will do

• Ship LLM-powered features with clear success metrics and failure handling.
• Build retrieval, tool-use, and workflow automation for domain-specific products.
• Create evaluation harnesses, guardrails, and monitoring for live AI systems.
• Work with product and design on useful, trustworthy AI-assisted experiences.

Responsibilities

• Prototype and productionize AI features from concept to rollout.
• Own model integration, prompt systems, and retrieval architecture.
• Monitor latency, cost, quality, and safety in production.
• Document patterns for reuse across ventures.

Requirements

• 3+ years in software engineering with meaningful applied AI work.
• Experience with Python, modern LLM APIs, and vector retrieval patterns.
• Understanding of eval design, prompt systems, and production failure modes.
• Bias toward shipping usable intelligence, not research-only experiments.

Nice to have

• Experience with RAG in regulated or enterprise contexts.
• Familiarity with fine-tuning, agents, or workflow orchestration tools.
• Exposure to EdTech, MediaTech, or operations-heavy products.

Benefits

• Real production AI problems across multiple venture domains.
• Room to define the AI layer for the portfolio over time.
• Direct collaboration with founders and product leaders.
• Compensation reflecting ownership depth across venture products.

Interview process

• Applied AI portfolio / work review
• System design for production LLM feature
• Hands-on problem solving
• Venture fit and communication round

Opens the application form in a new tab.