AI Platform Engineer

Harper•San Francisco, CA
•Onsite

About The Position

The AI Platform Engineer will architect and develop the core infrastructure powering an AI-enabled commercial insurance brokerage. The role spans the company’s proprietary AI Grid context engineering system, evaluation infrastructure, ML models for market-making and reasoning, data pipelines, and systems supporting AI agents across growth, sales, operations, and customer service. The engineer will build both ambient agents with background context and memory and the systems behind frontier agents and human-AI interfaces, with a focus on reliability, user experience, accuracy, and low latency. Working directly with the CEO and CTO, the engineer will build and ship production systems in a fast-moving, early-stage startup environment. The position involves rapid prototyping followed by production scaling, close collaboration with forward deployed engineers, and a strong bias toward action. The role is based in San Francisco and requires working in the office 5.5 days per week; relocation assistance is provided. The technology environment includes TypeScript/Node.js and Python; AI agent infrastructure such as AI Grid, Temporal.io, Pydantic-AI, and MCP servers; evaluation systems and human-annotated datasets for SFT and RL; event sourcing with Redis streams and PostgreSQL; Lambda architecture; RAG and vector databases; voice AI; modern data warehouse and ML platforms; orchestration and data integration tools; payment integrations; and observability and analytics systems.

Requirements

  • Strong experience with both TypeScript/Node.js and Python.
  • Deep understanding of distributed systems principles, CAP theorem tradeoffs, and event sourcing architecture.
  • Experience with modern data warehouse solutions such as Databricks, Astronomer, Snowflake, or similar.
  • Proven track record building ML modeling infrastructure and ETL/ELT pipelines.
  • Experience with evaluation systems and human-annotated datasets for ML training.
  • Experience building ML models for complex systems such as recommendation engines, market-making, or reasoning.
  • Experience with MCP server architecture and protocol implementation.
  • Experience with voice AI systems and conversational interfaces.
  • Experience with payment provider integrations and financial systems.
  • Experience designing data pipelines that serve multiple destination systems.
  • Experience with Temporal.io workflows or similar durable execution frameworks.
  • Experience with data enrichment and DAG-based orchestration.
  • Proven track record building production AI/ML systems at scale.
  • Experience with vector databases such as Qdrant, Pinecone, or Weaviate and with RAG systems.
  • Strong understanding of context engineering for AI systems.
  • Advanced usage of Cursor or WindSurf coding IDE.
  • Must be based in San Francisco and work in-office 5.5 days per week.

Nice To Haves

  • Exceptional ability in one or more of distributed systems, AI agents, context engineering, or data engineering.
  • Ideally, experience working in an early-stage startup environment and familiarity with its pace.

Responsibilities

  • Extend and enhance the proprietary AI Grid context engineering system, combining ETL with LLM pipelines and graphs.
  • Build evaluation systems using datasets and human-annotated data for supervised fine-tuning and reinforcement learning.
  • Develop ML models for underwriter load balancing, market-making, and reasoning systems.
  • Architect and maintain data pipelines that process multiple sources and deliver data to systems such as ClickHouse, vector databases, and ML platforms.
  • Build MCP servers that expose memory and tools to AI agents.
  • Create integrations with payment providers and financial systems.
  • Develop growth engineering infrastructure and agents that determine campaign strategies and optimize outreach.
  • Build voice AI systems for follow-ups, information collection, and outbound campaigns.
  • Create customer service AI infrastructure that enables one person to manage large numbers of leads and customers.
  • Design batch and real-time ETL/ELT pipelines and implement Lambda architecture patterns combining event streaming with batch processing.
  • Partner with forward deployed engineers to align platform capabilities with business needs.
  • Build observability systems to monitor agent reliability and performance.
  • Develop data enrichment pipelines and DAG-based orchestration systems.
  • Build and scale platform capabilities through prototyping, productionization, and iteration.
  • Fine-tune models using human-annotated datasets and integrate systems into a unified, observable platform.

Benefits

  • Relocation assistance; relocation costs will be covered.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service