Artificial Intelligence Engineer

Space Executive; New York (NY); On-site; USD 120,000 - 190,000; Full time

Job summary

Space Executive in New York seeks an Applied AI Engineer to own the document pipeline end-to-end, from ingestion to search, enabling AI systems to reason over high-quality data.

You'll work across infra to AI services, building machine-consumption search, shaping metadata and the knowledge graph, and delivering APIs. All levels considered; onsite five days.

Qualifications

• Buy-side context in asset management or hedge funds.
• Experience with search, retrieval, or document-processing systems.
• Comfort with Python and TypeScript; Kubernetes and Postgres; OpenSearch/Elasticsearch a plus.
• Experience with AI extraction/evaluation pipelines and structured outputs.
• Willing to work from infrastructure up to AI services layer.
• Adaptable in a fast-moving startup.

Responsibilities

• Own the document pipeline end to end from ingestion to search.
• Build search for machine consumption, shaping metadata and the knowledge graph around how an AI plans and executes multi-step work.
• Make quality measurable through extraction evals, precision and recall against labeled query sets, and automated checks that extracted fields match sources.
• Expose core capability as APIs, hardening the internal tooling layer so other systems can build against it.
• Take on customer-driven R&D above the data layer, including screening and analysis features.
• Work down the stack where the problem requires it, across integrations, backend services and the infrastructure they run on, with infrastructure changes shipped as code.

Skills

Python

TypeScript

FastAPI

Pydantic

SQLAlchemy

Kubernetes

Postgres

OpenSearch

Elasticsearch

Tools

Kubernetes

Postgres

OpenSearch

Elasticsearch

Job description

Applied AI Engineer, Search & Integration

New York, onsite 5 days | All levels considered

About the client:

An AI-native company whose software supports investment research at banks, asset managers and hedge funds. Their platform is in daily use by institutional research teams, and they're well funded and growing quickly after a recent round backed by venture investors and strategic partners from within financial services.

The role:

This team works across the entire product, from the public APIs down through the AI services layer to the data underneath it. You'd own the document pipeline end to end: ingestion, parsing, chunking, extraction, indexing and search. It's the foundation everything else depends on, since the quality of what the AI can reason over is set entirely by what this pipeline produces.

What you'll do:

• Own the pipeline end to end, from connectors through to search, where the bar for done is that the data layer genuinely serves the AI on top of it rather than merely working in isolation
• Build search designed for machine consumption rather than human browsing, shaping metadata, filters and the knowledge graph around how an AI plans and executes multi-step work
• Make quality measurable, through extraction evals, precision and recall against labeled query sets, and automated checks that extracted fields match their sources
• Expose core capability as APIs, hardening the internal tooling layer so other systems can build against it
• Take on customer-driven R&D above the data layer, including screening and analysis features
• Work down the stack where the problem requires it, across integrations, backend services and the infrastructure they run on, with infrastructure changes shipped as code

What they're looking for:

• Buy-side context: Engineering experience inside, or building for, an asset manager or hedge fund, with a real feel for how institutional research gets used day to day
• Search and retrieval: Production experience building or running search, retrieval or document-processing systems. The strongest candidates can describe the corpus they worked with, how its characteristics shaped their design, and how they knew whether search was actually performing
• Stack: Python (FastAPI, Pydantic, SQLAlchemy or similar) and comfort in TypeScript. Kubernetes, Postgres and a search engine such as OpenSearch or Elasticsearch are a plus
• AI systems: Multi-model extraction or evaluation pipelines, including retry and fallback design and structured outputs
• Range: Willing to work from infrastructure up to the AI services layer
• Adaptability: At ease in a fast-moving, technically deep startup