No Hallucination Guarantee is live. Learn more about the No Hallucination Guarantee

Precision at scale for public market research

The Hudson Labs Co-Analyst is AI designed for institutional workflows where accuracy is non-negotiable, relevance matters and long time-horizon context is crucial.

Anyone can say they're an expert. But only one of us wrote the textbook.

Hudson Labs has been built on language models from day one. In 2021 we launched the world's first financial software fully powered by LLMs.

Our CTO and co-founder Suhas Pai authored Designing Large Language Model Applications, published by O'Reilly Media in March 2025. It has been translated into more than 10 languages and is used by AI engineers around the world.

Suhas Pai, CTO and Co-Founder of Hudson Labs

Suhas Pai

CTO and Co-Founder, Hudson Labs

Designing Large Language Model Applications, O'Reilly book by Suhas Pai

Core capabilities

01

Multi-period financial precision

Financial research requires consistency across time. Most competitive tools rely on generalist models via ChatGPT-style wrappers and face similar accuracy limitations. Hudson Labs maintains accuracy across multi-document, multi-period financial queries.

02

Guidance identification

Subtle guidance cues are frequently missed by generalist AI systems. Common failure modes include missing forward-looking statements, confusing historical/current/future tense, and losing conditional guidance in longer responses. Hudson Labs uses specialized models and task-specific logic to consistently surface forward-looking statements.

03

Materiality (relevance) ranking

Correct information isn't always useful information. As document volume increases, irrelevance becomes as costly as hallucination. Hudson Labs applies proprietary relevance ranking to reduce noise, prioritize information tied to research questions, and preserve full traceability to source.

04

Verbatim, source-faithful quotes

Generalist AI systems frequently paraphrase while presenting text as quoted or surface language that doesn't exist in source material. Hudson Labs retrieves verbatim excerpts only, with direct linkage to underlying documents.

Why Hudson Labs works

01

AI systems architecture over AI models

Building effective AI products is far more about systems, pipelines, and architecture than models. This is achieved through context engineering optimized for long-horizon tasks, topic spans, unique pre-processing and meta-tagging, proprietary retrieval systems, and task-specific system design.

02

Deep, internal AI expertise

Hudson Labs is AI-native and builds its AI in house. It does not outsource research or build thin wrappers around generalist models. The CTO and co-founder authored a bestselling O'Reilly textbook on Large Language Models, widely used by engineers building real-world AI systems.

03

LLM-powered history

Hudson Labs launched the first fully LLM-powered financial services software application in 2021, when large language models were still largely experimental. Proprietary relevance ranking developed during that period remains a core platform component.

Hudson Labs vs. general AI models

The Co-Analyst extracts detailed quantitative and qualitative financial data across multiple documents and long time periods, with nearly 100% accuracy and zero hallucination.

On the Vals AI financial retrieval benchmark, leading general-purpose models (GPT-5.2, GPT-5.1, Opus 4.5) score 53–57% in accuracy on financial tasks involving retrieval and analysis of specific financial information.

Source: Vals AI benchmark, December 2025

Design principles

01

Automate the grunt work and leave judgment to analysts

We automate the tedious parts of financial research (data gathering, cross-referencing and formatting) so analysts can focus on judgment and decision-making.

02

Narrow and effective systems outperform complex, error-prone ones

Rather than building a general-purpose AI, we build purpose-specific models trained on financial documents. Narrow scope means higher precision.

03

Every output must have a direct and traceable connection to source material

If you cannot verify it, it should not inform a trade. Full transparency is non-negotiable.

Try the Co-Analyst free

Use it for 14 days and see how purpose-built financial AI compares to general-purpose tools.

View products