AI Systems Engineer for Enterprise Safety Platform

Aifund; United States; On-site; USD 180,000 - 300,000; Full time

Job summary

Haven Safety invites an experienced AI/LLM engineer to join the founding team in the US. You will design, ship, instrument, and improve multi-step reasoning pipelines that connect Haven's evidence base and knowledge graph to investigator-facing product experiences.

This role reports to the CTO and demands precision, traceability, and strong production discipline. You will own production LLM systems, manage tool calls and agent orchestration, and work across OpenAI, Anthropic, and Google.

Qualifications

• Build and operate multi-step LLM pipelines coordinating model calls, tool calls, graph queries and retrieval.
• Design context layers across graph traversal and hybrid retrieval to feed models with right evidence.
• Establish datasets, regression suites, and human-review loops for quality attribution.

Responsibilities

• Production reasoning systems: build and operate multi-step LLM pipelines.
• Agent orchestration: extend the coordinated agent team and orchestration layer.
• Grounding and retrieval: ensure right evidence and knowledge are retrieved for each call.
• Evaluation: create scoring, regression tests, and human-label loops.
• Production AI operations: implement tracing, audits, monitoring, and dashboards.
• Technical direction: select models and shape the AI roadmap with product teams.
• Production LLM systems: ship reliable LLM-powered features used by customers.
• Agentic workflows: build and debug multi-step tool-calling workflows.
• Evaluation discipline: establish repeatable LLM evaluation across datasets.
• Retrieval judgment: decide what context to retrieve and how much.
• Production ownership: own deployment through monitoring and incident response.
• Model judgment: explain tradeoffs across quality, latency, cost, and risk.
• Graph reasoning: work with Neo4j/Cypher and live knowledge graphs.
• Security: guard against prompt injection and data leakage.
• Azure and hybrid search: operate on Azure and similar platforms.

Skills

Python

Neo4j

Cypher

FastAPI

LangChain

LangGraph

Azure

OpenAI

Anthropic

Google

Tools

Neo4j

Cypher

Azure AI Search

Pinecone

MongoDB Atlas

Job description

Haven Safety invites an experienced AI/LLM engineer to join the founding team in the US. You will design, ship, instrument, and improve multi-step reasoning pipelines that connect Haven's evidence base and knowledge graph to investigator-facing product experiences.

This role reports to the CTO and demands precision, traceability, and strong production discipline. You will own production LLM systems, manage tool calls and agent orchestration, and work across OpenAI, Anthropic, and Google.