UnfoldIPThemes

Company patents

HUBSPOT, INC.

HUBSPOT, INC.'s patent strategy reveals a strong, dual focus on "Business Methods & Fintech" and "Databases & Information Retrieval," both comprising 57.4% of its portfolio, with a surprising surge in 2025 where both categories saw over 350% year-over-year growth. While "Machine Learning & AI" also represents a significant portion (48.1%) and experienced a 200% growth in 2025, the sharp decline in patenting activity across nearly all categories so far in 2026, including a 76.5% drop in "Business Methods & Fintech" and a 77.8% drop in "Databases & Information Retrieval," suggests a potential shift in patenting cadence or a strategic re-evaluation following the aggressive filing year of 2025.

Patent Trend by Technology Area

Yearly patent publications since 2023

Product themes

Product-level themes inferred from filings since 2023, with category chips showing where each theme appears. Select a theme to filter the patents below.

54 US filings (since 2023) · 9 categories · 10 themes

Specialized Data Integration

Methods and systems for integrating, transforming, and managing complex or domain-specific data from disparate sources into a unified structure, often for specific applications like social networks, genomics, or business forms.

Databases & Information Retrieval
Who else files here? →
28since 2023
+400.0%YoY
Dynamic Content & Engagement Features

Features within messaging platforms that enhance user interaction and content consumption through intelligent suggestions, content persistence mechanisms, engagement analytics, and adaptive presentation of conversational media.

Messaging & Email
Who else files here? →
15since 2023
+100.0%YoY
Secure & Context-Aware Messaging

Methods and systems for enhancing the security and privacy of electronic messages, often by integrating contextual data such as location, social network graphs, or user authentication levels to control access, filter content, or enable specific group interactions.

Messaging & Email
Who else files here? →
8since 2023
+500.0%YoY
Message System Management & Delivery

Core infrastructure and operational techniques for efficient and reliable message handling, including server-side logic for managing subscriptions, aggregating messages, optimizing network connections, and ensuring data consistency across distributed messaging services.

Messaging & Email
Who else files here? →
7since 2023
+400.0%YoY
Automated Transaction Systems

Systems designed to streamline and automate various commercial transactions, including mobile-enhanced processes, secure online checkouts, customer service interactions, and privilege issuance, often leveraging digital authentication.

Business Methods & Fintech
Who else files here? →
7since 2023
new
Personalized Recommendations

Systems that use user data, preferences, and machine learning to generate tailored advice, product recommendations, goal-setting plans, or contextual information for individuals across different domains.

Business Methods & Fintech
Who else files here? →
3since 2023
new
Context-Aware Conversational AI

AI systems designed to engage in natural language dialogue, maintain conversation state, understand user intent, and generate relevant responses, often across multiple communication channels or modalities.

Speech Processing
Who else files here? →
3since 2023
+100.0%YoY
Multimodal Data Fusion

Techniques for combining and analyzing information from multiple distinct data modalities (e.g., text, image, video, audio, sensor data) to derive richer insights or improve system performance and decision-making.

Machine Learning & AI
Who else files here? →
2since 2023
+100.0%YoY
Federated & Distributed ML

Methods for training machine learning models across multiple decentralized devices or servers while keeping data localized, often involving aggregation of model parameters and secure communication.

Machine Learning & AI
Who else files here? →
2since 2023
n/a
MLOps & Model Deploymentfiltered

Systems and methods for automating the lifecycle of machine learning models, including pipeline deployment, model management, versioning, and configuring for different inference environments.

Machine Learning & AI
Who else files here? →
2since 2023
n/a

Patents

Showing 1-4 of 4

MLOps & Model Deployment
Page 1 of 1
US 20230351227 A1APPLICATION
G06N5/04

MULTI-CLIENT SERVICE SYSTEM PLATFORM

Filed:2023-07-02Pub:2023-11-02
Applicant:HubSpot, Inc.

A modular machine learning-as-a-service (MLAAS) system uses machine learning to respond to tasks without requiring machine learning modeling or design knowledge by its users. The MLAAS system receives an inference request including a model identifier and a target defining features for use in processing the inference request. The features correspond to a task for evaluation using a machine learning model associated with the model identifier. An inference outcome is generated by processing the inference request using the target as input to the model. Feedback indicating an accuracy of the inference outcome with respect to the task is later received and used to generate a training data set, which the MLAAS can use to further train model used to generate the inference outcome. As a result, the training of a machine learning model by the MLAAS system is limited to using data resulting from an inference performed using that model.

US 11727287 B2GRANTED
G06N5/04

Multi-client service system platform

Filed:2022-08-08Pub:2023-08-15
Applicant:HubSpot, Inc.

A modular machine learning-as-a-service (MLAAS) system uses machine learning to respond to tasks without requiring machine learning modeling or design knowledge by its users. The MLAAS system receives an inference request including a model identifier and a target defining features for use in processing the inference request. The features correspond to a task for evaluation using a machine learning model associated with the model identifier. An inference outcome is generated by processing the inference request using the target as input to the model. Feedback indicating an accuracy of the inference outcome with respect to the task is later received and used to generate a training data set, which the MLAAS can use to further train model used to generate the inference outcome. As a result, the training of a machine learning model by the MLAAS system is limited to using data resulting from an inference performed using that model.

US 20220383199 A1APPLICATION
G06N20/00

MULTI-CLIENT SERVICE SYSTEM PLATFORM

Filed:2022-08-08Pub:2022-12-01
Applicant:HubSpot, Inc.

A modular machine learning-as-a-service (MLAAS) system uses machine learning to respond to tasks without requiring machine learning modeling or design knowledge by its users. The MLAAS system receives an inference request including a model identifier and a target defining features for use in processing the inference request. The features correspond to a task for evaluation using a machine learning model associated with the model identifier. An inference outcome is generated by processing the inference request using the target as input to the model. Feedback indicating an accuracy of the inference outcome with respect to the task is later received and used to generate a training data set, which the MLAAS can use to further train model used to generate the inference outcome. As a result, the training of a machine learning model by the MLAAS system is limited to using data resulting from an inference performed using that model.

US 11449775 B2GRANTED
G06N5/04

Multi-client service system platform

Filed:2019-12-17Pub:2022-09-20
Applicant:HubSpot, Inc.

A modular machine learning-as-a-service (MLAAS) system uses machine learning to respond to tasks without requiring machine learning modeling or design knowledge by its users. The MLAAS system receives an inference request including a model identifier and a target defining features for use in processing the inference request. The features correspond to a task for evaluation using a machine learning model associated with the model identifier. An inference outcome is generated by processing the inference request using the target as input to the model. Feedback indicating an accuracy of the inference outcome with respect to the task is later received and used to generate a training data set, which the MLAAS can use to further train model used to generate the inference outcome. As a result, the training of a machine learning model by the MLAAS system is limited to using data resulting from an inference performed using that model.

1