Docyt has launched a new AI engine, dubbed High Precision Accounting Intelligence (HpAI), aimed at automating intricate accounting processes.

This development follows the company’s recent pre-Series B funding round, which secured $12m from Pivot Investment Partners and other existing investors.

The funding is intended to enhance the deployment of Docyt’s AI solutions within accounting firms and multi-entity organisations.

The HpAI engine combines large language models with specialised AI designed for complex tasks such as reconciliation, categorisation, anomaly detection, and month-end closing.

The capabilities of HpAI are integrated into the Docyt Accountant Copilot, which features AI agents that mimic the expertise of senior bookkeepers.

This allows firms to transfer responsibilities typically assigned to junior accountants.

Docyt’s AI system is capable of processing data, generating analytics, identifying anomalies, and automating month-end closing procedures. By using the Copilot, firms can potentially increase their operational efficiency without the need for additional personnel.

The HpAI technology is based on foundational models trained on comprehensive bookkeeping data, including ERP charts of accounts, vendor information, bank transactions, and financial records.

The models utilise 128 billion data points across more than 20 industry sectors, supported by a structured bookkeeping dataset developed over five years.

According to Docyt, these models do not require retraining on individual customer data, as they can adapt in real-time by maintaining contextual information to ensure accuracy.

This approach contrasts with commercial models that depend on broader datasets and general fine-tuning, aiming instead for precise categorisation and relevant financial insights.

Additionally, Docyt has indicated that its automation capabilities cover core bookkeeping functions and are applicable to various sectors, including hospitality, with plans for further expansion into additional industries.

Docyt co-founder and CEO Sid Saxena said: “At Docyt, we have spent over five years building high-quality synthetic datasets using in-house expertise of expert accountants doing high-quality data labeling that captures the complexity of real-world bookkeeping.

“This dataset is a critical component of our HpAI architecture, enabling it to deliver precise, context-aware automation. “We are now bringing this foundational AI architecture to accounting firms in the form of Docyt AI Copilot, enabling them to deliver client bookkeeping at scale, and with precision.”