Artificial Intelligence Engineer

Space Executive; New York (NY); On-site; USD 180,000 - 280,000; Full time

Job summary

Space Executive in New York seeks an experienced engineer to own the document pipeline from ingestion to indexing and search, empowering an AI research agent to reason over institutional materials.

You will design agent-focused search, convert products into usable APIs, and lead customer-driven R&D such as screening and portfolio-analysis features, all within a fast-moving startup environment.

Qualifications

• Proficient in building production-grade data/document pipelines.
• Experience with AI/LLM extraction pipelines and evaluation.
• Comfortable operating in a fast-moving startup environment.

Responsibilities

• Own the document pipeline end to end (connectors, parsing, chunking, extraction, indexing, search).
• Design search for agents and metadata/filters to support multi-step planning.
• Turn core product into APIs and micro-services for external consumption.
• Build customer-driven R&D such as quantitative screening and portfolio-analysis features.
• Work across infrastructure, backend services, and the cloud platform.

Skills

Python (FastAPI, Pydantic, SQLModel/SA

TypeScript

Full-Stack engineering

AI/LLM systems

Adaptability

Tools

Kubernetes

OpenSearch

Postgres

Terraform

Helm

ArgoCD

SQL

Job description

New York | All levels considered (new grad through senior)

About the Company:

Our client is an AI-native company building the intelligence layer for global public markets. Their platform is used daily by 70+ financial institutions across the US, Europe, and Asia, spanning sell-side research teams and buy-side asset managers and hedge funds collectively managing over $5 trillion in assets. They recently closed a $22M Series A, backed by a global syndicate of strategic financial institutions and venture investors.

The Opportunity:

The Applied AI Engineering (Search & Integration) team cuts across the whole product, from APIs through the AI services (search, chat, the autonomous research agent) down to the data layer beneath them. This role owns the document pipeline end to end, from ingestion through parsing, chunking, extraction, indexing, and search, and is central to how the company's AI research agent finds and reasons over institutional research material.

What You’ll Do:

• Own the document pipeline end to end (connectors, parsing, chunking, extraction, indexing, search), with "the data layer actually serves the research agent" as the definition of done
• Design search for agents, not just people, shaping the identification layer (metadata, filters, knowledge-graph evolution) around how the research agent plans multi-step work
• Turn the core product into APIs, hardening the internal skill/tool system for external systems to consume
• Build customer-driven R&D on top of the core data layer, such as quantitative screening and portfolio-analysis features
• Go down the stack when needed, including integrations, connectors, backend services, and the infrastructure they run on
• Own cloud infrastructure decisions directly, with changes shipped as pull requests against Terraform/Helm/ArgoCD
• Make improvements measurable through extraction-quality evals, search precision/recall against labeled query sets, and field-level parity checks

What They’re Looking For:

• Buy-Side Background: Engineering experience within, or built for, an asset manager or hedge fund, with a real understanding of how institutional research actually gets consumed
• Search & Retrieval: Hands-on experience building or operating search, retrieval, or document-processing systems in production. Strong candidates can speak to the nature of the document corpus they worked with, how it shaped their search design, and how they measured whether search actually performed, and why
• Engineering Stack: Python (FastAPI, Pydantic, SQLModel/SQLAlchemy) and comfort with TypeScript; exposure to Kubernetes, OpenSearch, and Postgres is a plus
• AI/LLM Systems: Experience with multi-model LLM extraction or evaluation pipelines, including retry/fallback design and structured outputs
• Full-Stack Comfort: Willingness to go down the stack, from infrastructure through to the AI services layer
• Adaptability: Comfortable operating in a fast-moving, technically deep startup environment

Logistics:

• Location: New York (Onsite 5 days per week)
• Stage: Series A ($22M, closed July 2026)