Graduate AI / ML Engineer (RAG, Chatbot, LLM, Vector DB)

Requirements

1–3 years of experience in ML, data science, or backend engineering

Strong Python skills

Experience building APIs or backend systems

Experience with machine learning modeling (e.g., regression, time-series, classification, or similar)

Exposure to LLMs, chatbots, or prompt engineering

Comfortable working with messy datasets

Nice to Have: RAG or vector search experience; Time-series forecasting (Prophet, XGBoost, etc.); Retail / supply chain data experience; MLOps or production ML exposure

Why Join: Build real AI products (not just models); Work on LLMs, chatbots, and predictive ML systems; High ownership and fast growth; Be part of a major platform rebuild

Compensation: Competitive salary; Health benefits; Hybrid work model

Optional (but high leverage): Please include examples of ML models or LLM projects you’ve built (GitHub or portfolio).

Projects

Krunchbox is a retail analytics SaaS platform helping brands increase sell-through, prevent stockouts, and uncover lost revenue across major retailers. We’re launching Krunchbox Reimagined — a modern AI platform focused on predictive analytics, AI agents, and real-time decision support. We’re looking for an AI / ML Engineer (1–5 years experience) to help us build chatbots, RAG systems, and production-grade ML models used directly by customers. This role spans building AI agents that generate insights, reports, and recommendations; developing RAG pipelines blending LLMs with structured data (POS, inventory, product data); and creating chat-based experiences for customer analytics.

What You’ll Do

• Build AI agents (“Krunchy”) that generate Insights, Reports, Recommendations
• Develop RAG pipelines combining LLMs with structured data (POS, inventory, product data)
• Create chat-based experiences for customer analytics
• Machine Learning Modeling (Core): build and improve models for Demand forecasting, Stockout risk, Lost sales estimation, Anomaly detection
• Perform feature engineering on messy retail datasets
• Model evaluation and iteration
• Help take models from prototype → production
• Tech Stack: Python (FastAPI preferred), LLM APIs (OpenAI, Anthropic), LangChain / LlamaIndex (or similar), Vector databases, ClickHouse / modern data stack, AWS / cloud infrastructure

Benefits

• Competitive compensation package.
• Comprehensive health and benefits coverage.
• A predominantly in-person, collaborative work environment located in Santiago to encourage fast iteration and real-time problem solving.
• Opportunity to scale and lead a global SaaS platform that solves real-world customer challenges.
• A direct, impactful role in shaping the future of AI-powered supplier-retailer collaboration.

Nice to Have

• RAG or vector search experience
• Time-series forecasting (Prophet, XGBoost, etc.)
• Retail / supply chain data experience
• MLOps or production ML exposure