Daybreak

About the Role We are seeking an AI Engineer to architect and optimize our agent-based forecasting and decision systems. You'll work at the intersection of LLMs, distributed computing, and statistical forecasting to build intelligent agents that transform how enterprises plan their supply chains. This role combines hands-on development of AI agents with rigorous evaluation frameworks to ensure our systems deliver measurably superior business outcomes. Core Responsibilities Agent Development & Context Engineering Design and implement intelligent agents using Pydantic AI and similar frameworks for demand planning and decision-making Craft sophisticated system prompts and agent instructions that encode domain expertise and business heuristics Test & implement context pipeline improvements to optimize agent performance and cost Build evaluation frameworks to measure agent performance against statistical baselines (MAPE, WMAPE, bias metrics) Implement self-improving agent systems that learn from feedback and adapt to customer-specific patterns Data Engineering & Analytics Build scalable data pipelines using Databricks/Spark for processing millions of SKU-location combinations Optimize distributed computing patterns to minimize costs while maximizing performance Design feature engineering strategies for time-series forecasting and anomaly detection Create data quality validation frameworks for ML model inputs and outputs LLM Integration & Workflow Automation Develop multi-agent orchestrations for complex supply chain workflows Integrate code execution environments with LLMs for autonomous data analysis Build custom tools and function calling patterns for agent-based systems Design fallback strategies and error handling for production agent deployments Experimentation & Evaluation Design A/B testing frameworks for comparing agent strategies against traditional methods Build comprehensive evaluation suites measuring both accuracy and business impact Create simulation environments for testing agent behaviors under various scenarios Develop metrics dashboards showing per-customer optimization gains Required Qualifications 3+ years of experience in data science, ML engineering, or analytics roles Strong Python skills with expertise in data manipulation (Pandas, PySpark, Polars) Experience with SQL and distributed computing frameworks (Spark/Databricks) Experience with cloud deployments and IaC Hands-on experience with LLMs and prompt engineering Background in statistical analysis and experiment design Experience building data pipelines and ETL processes Preferred Qualifications Experience with supply chain analytics or demand forecasting Knowledge of agent frameworks (Pydantic AI, LangChain, AutoGPT) Familiarity with time-series forecasting methods (ARIMA, Prophet, neural approaches) Experience with workflow orchestration tools (Dagster, Airflow) Background in optimization algorithms or operations research Previous work with enterprise B2B customers Technical Environment AI/Agent Stack: Pydantic AI, Claude/GPT-4, custom evaluation frameworks Data Platform: Databricks, Spark Connect, Dagster, Airbyte ML Libraries: XGBoost, LightGBM, statsmodels, scikit-learn, transformers Infrastructure: Kubernetes, Docker, AWS, PostgreSQL Languages: Python (primary), SQL, potentially TypeScript for UI integrations What Makes This Role Unique You'll pioneer the application of agentic AI to supply chain optimization, where small accuracy improvements translate to millions in inventory savings. This isn't about building chatbots – it's about creating autonomous systems that outperform traditional forecasting methods through intelligent reasoning and adaptive learning. You'll have direct impact on how Fortune 500 companies plan their operations. Growth Opportunities Drive the development of our next-generation agent architecture Publish research on agent-based forecasting methods Present at conferences on practical applications of LLMs in enterprise Shape the technical direction of our AI strategy Mentor teams on agent development best practices Ideal Candidate Profile You're someone who gets excited about turning messy real-world data into actionable insights through AI. You understand that in enterprise B2B, the "analysis battle" matters more than pure accuracy – customers need to understand and trust AI recommendations. You're comfortable with ambiguity, enjoy experimenting with new approaches, and have the resilience to iterate through multiple solutions. Most importantly, you're customer-obsessed and understand that each enterprise has unique patterns that require tailored optimization strategies. We offer competitive compensation, equity participation, and the opportunity to define how AI transforms supply chain management at scale. Daybreak is an Equal Opportunity Employer