Staff Software Engineer – AI Retrieval & Search Systems

Helios Intelligence Platforms; New York (NY); On-site; USD 150,000 - 230,000; Full time

Job summary

Helios Intelligence Platforms is hiring a Staff/Senior Software Engineer to own Proxi’s retrieval stack end to end, from corpus onboarding and index design through hybrid retrieval, fusion, reranking, reindexing, and backfills. You will work with the CTO and founding team in a fast-moving, high-stakes environment in New York City.

You will advance query understanding, temporal-retrieval capabilities, and cross-collection search, while optimizing performance, reliability, and cost.

Qualifications

• Experience with retrieval systems and search architectures.
• Knowledge of lexical and vector-based indexing techniques.
• Familiarity with multilingual and cross-lingual retrieval is a plus.

Responsibilities

• Own Proxi’s retrieval stack end to end from onboarding to backfills.
• Advance query understanding and temporal retrieval capabilities.
• Improve retrieval performance, reliability, and cost via caching and optimization.
• Expand federated search across Proxi’s data collections and refine result merging.
• Enhance time-aware retrieval to access correct historical states.
• Map documents to people, orgs, events, and relationships.
• Support multimodal content search within the same interface.
• Own relevance quality and production observability and recovery.

Skills

Retrieval systems

BM25 parameterization

HNSW indexing

Learning to rank

Entity linking

Temporal modeling

PII filtering

Multilingual retrieval

Tools

HNSW parameters

Vector indices

Job description

Helios Intelligence Platforms is hiring a Staff/Senior Software Engineer to own Proxi’s retrieval stack end to end, from corpus onboarding and index design through hybrid retrieval, fusion, reranking, reindexing, and backfills. You will work with the CTO and founding team in a fast-moving, high-stakes environment in New York City.

You will advance query understanding, temporal-retrieval capabilities, and cross-collection search, while optimizing performance, reliability, and cost.