Applied AI Engineer

About the role

Block Labs is building governed AI agents that interact with customers and internal teams. As an Applied AI Engineer you will create production‑grade agents, models and dashboards that drive real‑time decisioning and risk management.

Key responsibilities

• Design and ship Slack‑native analyst agents that translate natural‑language queries into governed SQL, validate results and provide evidence‑backed answers.
• Develop customer‑facing agents with retrieval‑augmented generation, multi‑turn conversation flows, escalation logic and strict security controls.
• Build and productionise ML models for churn, lifetime value, risk scoring, fraud detection and anomaly detection, serving them via Kafka, ClickHouse and batch pipelines.
• Create audit‑ready dashboards and approval surfaces that expose agent decisions, model outputs and tool calls to business stakeholders.

Required profile

• 4+ years of software, data‑science or ML engineering experience, including at least 1 year building LLM‑powered agents in production.
• Hands‑on experience with retrieval‑augmented generation, grounding, hallucination control and prompt‑injection defence.
• Proven ability to own the full ML lifecycle: feature engineering, training, serving, monitoring and automated retraining.
• Strong statistical rigour – experiment design, hold‑outs, uplift measurement and score calibration.

Required skills

• Python for production services
• TypeScript for front‑end review and approval interfaces
• SQL on columnar analytical databases (ClickHouse preferred)
• Kafka/MSK for real‑time model serving
• LangGraph or Anthropic Agent SDK for multi‑step LLM orchestration
• RAG techniques, retrieval, memory management and LLM prompting
• Machine learning and data‑science tooling