Blob Hub

What Blob Hub does

Blob Hub (operating as BlobHub) is a private, self-serve cloud platform founded in 2019 and headquartered in Redmond, Washington, that targets developers and AI/ML practitioners building agentic applications. The platform combines a visual Workflow Editor, a step-by-step Workflow Debugger, and a customizable Workflow Playground for creating, inspecting, and experimenting with generative AI and agent-based workflows. Workflows are assembled from building blocks that connect to leading third-party AI providers — OpenAI, Anthropic Claude, and Google Gemini — alongside Pinecone for vector retrieval and Firecrawl and Tavily for web scraping and search, with custom code blocks for extension. The platform also exposes a Python SDK via PyPI and a REST API for programmatic control.

Underlying the workflow surface, BlobHub operates managed data stores spanning graph data (linked data repositories with query, visualization, and embeddable widgets), datasets (including Kensho Derived Wikimedia with 75M+ items, ConceptNet 5.7.0, and SuperGLUE benchmarks), structured relational blobs, and strongly versioned Managed ONNX Blobs for deep neural network models with permissions and preview. The technology stack is anchored on a custom Workflow BlobHub interpreter and OrientDB SQL for graph queries.

Commercially, BlobHub pursues a product-led growth motion through free public access to its workflow gallery, documentation site, example notebooks, and SDK distribution, with premium and enterprise pricing offered on a quote basis. Revenue is not publicly disclosed, no funding rounds or institutional investors are reported, and only a single technologist (Egor Pushkin) is identified in management. The company serves a global developer base with no stated geographic restrictions.

Name

Blob Hub

Legal name

BlobHub

Website

https://blobhub.io

Company type

Private

Founded year

2019

Operating status

Operating

Short description

Blob Hub is a self-serve cloud platform for building, debugging, and deploying agentic AI workflows, serving developers and AI/ML practitioners with visual workflow tools, managed ONNX model storage, and managed graph and dataset stores, monetized via freemium and quote-based premium tiers.

Ownership category

akta.pro rank

Product category

AI Workflow Development Platform

NAICS

Computer Systems Design and Related Services (5415), Computer Systems Design and Related Services (54151), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (518210)

SIC

Services-Prepackaged Software (7372), Services-Computer Programming, Data Processing, Etc. (7370)

akta.pro primary industry

AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs) (HDAEANAJ)

akta.pro secondary industries

iBPMS (Intelligent BPM Suites) & Process Automation Suites (HDAEAHAG), Model Development & Training Platforms (AutoML, Notebooks, Feature Stores) (HDAEANAB)

Keywords

• Agentic application development
• AI workflow development
• ONNX model versioning
• Graph data management
• Machine learning operations

Headquarters

HQ city

Redmond

HQ country

United States

HQ region

North America

Markets served

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Infrastructure, Personnel, Operations, Marketing or Sales

Revenue model

1. Platform Services: BlobHub generates revenue through its cloud-based platform providing workflow creation, debugging, deployment tools, and managed data storage for AI applications. The platform offers free tiers with premium features and usage-based pricing for advanced capabilities.

Pricing tiers

Go-to-market motion1 record

Distribution channels1 record

Marketing channels4 records

Core offering

BlobHub is a self-serve cloud platform for agentic application development that provides visual tools (Workflow Editor, Workflow Debugger, Workflow Playground) for building, debugging, and deploying generative AI and agent-based workflows. The platform combines these workflow tools with managed data stores — Graph Data, Datasets, Structured Data, and versioned ONNX Models — and exposes native integrations with leading AI model providers (OpenAI, Anthropic Claude, Google Gemini), vector databases (Pinecone), and web interaction tools (Firecrawl, Tavily) to support RAG, computer vision, and NLP applications. Programmatic access is available through a public REST API and a Python SDK distributed via PyPI.

Product overview

BlobHub is a platform for agentic application development that provides a unified environment for creating, debugging, and deploying generative AI and agent-based workflows. The platform consists of workflow management tools (Workflow Editor, Workflow Debugger, Workflow Playground) combined with managed data stores (Graph Data, Datasets, Structured Data, ONNX Blobs). Users build workflows using visual building blocks that connect to leading AI models (OpenAI, Anthropic, Gemini), vector databases (Pinecone), and web tools (Firecrawl, Tavily). The platform also offers pre-built ONNX models for computer vision (Super Resolution, Object Detection) and NLP (BiDAF), along with public datasets including SuperGLUE benchmarks and knowledge graphs. Python SDK and REST API enable programmatic access to all platform capabilities.

Differentiator

Problem solved

Functional benefit

Products and services

• Workflow Editor Visual workflow editor providing building blocks for creating generative AI and agent-based workflows with connectors to leading AI models and vector stores, built-in web interaction tools, and custom code block support, targeted at developers building AI-powered applications and agents.
• Workflow Debugger Visual debugging tool providing step-by-step workflow execution insights for analysis and troubleshooting during development and production phases, targeted at developers and AI/ML practitioners.
• Workflow Playground Customizable experimentation environment with plug-and-play interface for testing workflows during development, featuring widgets for workflow interaction and session data visualization, targeted at developers building AI-powered applications.
• Graph Data Store Managed linked data repositories with collaboration and publishing capabilities, providing tools for data exploration, visualization, ad-hoc OrientDB SQL querying, and flexible appearance templating for external embedding, targeted at developers and researchers working with knowledge graphs.
• Datasets Store Repositories of data in native formats ready for loading, integration, and consumption in applications, including NLP datasets such as SuperGLUE benchmarks, targeted at developers and AI researchers.
• Structured Data Store Traditional relational storage for general-purpose use cases with the ability to quickly ingest and share structured blobs of data, targeted at developers integrating structured data into applications.
• Managed ONNX Blobs Strongly versioned storage for deep neural network models with a centralized repository, flexible permission system, and native interface for model operations including organization, blob creation, access management, revisions, upload, download, and preview, targeted at data scientists and ML engineers running model operations.
• Python SDK Python software development kit built on top of the BlobHub REST API, enabling the full set of operations over ONNX blobs and supporting model management and inference workflows, distributed via PyPI and targeted at developers and data scientists.
• ONNX Vision Models Collection of pre-trained vision models in ONNX format including Object Detection (identifies individual objects in images) and Super Resolution (increases image resolution via efficient sub-pixel convolution neural network), targeted at developers building computer vision applications.
• ONNX Text Models Collection of pre-trained text models in ONNX format including the Bi-directional Attention Flow (BiDAF) network for extractive question answering and reading comprehension, targeted at developers building NLP applications.
• World Data Blobs Graph blobs populated with queryable data from various domains of human knowledge including WikiMedia (Kensho Derived Wikimedia Dataset with 75+ million items), Movies, and the ConceptNet 5.7.0 multilingual knowledge graph, targeted at developers and researchers building knowledge graph applications.
• SuperGLUE Datasets Natively-visualized NLP datasets covering SuperGLUE benchmark tasks (AX-b, AX-g, BoolQ, CB, COPA, MultiRC, ReCoRD, RTE, WiC, WSC) for language understanding evaluation, targeted at AI researchers and NLP practitioners.
• Deep Research Workflow Generative AI workflow template tailored for deep research applications, part of the Featured Blobs workflow gallery, usable as a starting point for building research assistants on BlobHub.
• Discussion Board Workflow Generative AI workflow template for discussion board applications, part of the Featured Blobs workflow gallery, originally created by egorpushkin about one year prior to the source capture.
• Open Canvas Workflow Generative AI workflow template for open canvas applications, part of the Featured Blobs workflow gallery, usable as a starting point for building open canvas experiences on BlobHub.
• Basic RAG Workflow Basic Retrieval-Augmented Generation pattern workflow that ingests web content into Pinecone and retrieves/summarizes relevant data, demonstrating the platform's ingest-web and retrieve flows.

Segments3 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration6 records

AI capability9 records

Feature9 records

Partnerships

Five partnerships are on record, tiered core.

• AnthropiccoreTechnology or IntegrationNative integration with Anthropic Chat interface for Claude-powered conversational AI workflows.
• Google GeminicoreTechnology or IntegrationNative integration with Gemini Chat interface for multi-turn messaging using Google's Gemini AI models.
• PineconecoreTechnology or IntegrationVector database integration for retrieval-augmented generation (RAG) workflows. Supports text chunk ingestion and retrieval operations.
• FirecrawlcoreTechnology or IntegrationWeb scraping API integration for ingesting web content into AI workflows. Used for RAG ingestion flows.
• ONNXcoreTechnology or IntegrationNative support for ONNX (Open Neural Network Exchange) format. Provides versioned storage for deep neural network models with centralized repository and permission system.

Scale indicators3 records

Recent moves6 records

Expansion highlights5 records

Direct peers

• LangChain: LangChain is the dominant open framework for building LLM-powered applications and agents with composable chains, retrievers, and tool integrations. BlobHub's visual workflow editor with multi-model chat blocks and RAG templates competes directly with LangChain's chain/agent abstractions for the same developer audience.
• Flowise: Flowise is an open-source drag-and-drop UI builder for LLM workflows and agents, with native connectors for OpenAI, Anthropic, vector stores, and document loaders. Its visual building-block approach mirrors BlobHub's Workflow Editor and serves the same 'developers who don't want to code chains by hand' use case.
• LlamaIndex: LlamaIndex is a leading data framework for connecting LLMs to private and external data, with strong RAG and agent tooling. BlobHub's Basic RAG workflow, Pinecone integration, and ingest/retrieve flow blocks target the same RAG developer workflow that LlamaIndex pioneered.
• Dust: Dust is a platform for designing, deploying, and orchestrating AI agents and assistants for teams, with model-agnostic LLM access and tool integrations. It overlaps directly with BlobHub's agent-building blocks, OpenAI Agents integration, and multi-model chat primitives.
• Stack AI: Stack AI is a no-code platform for building and deploying AI agents and workflows aimed at both developers and business users. Its drag-and-drop builder with RAG, chat, and tool integrations competes with BlobHub's Workflow Editor for the same agentic application audience.
• n8n: n8n is a source-available workflow automation platform that has added native LLM and AI agent nodes. Its visual workflow canvas and self-hostable/cloud model parallel BlobHub's Workflow Editor, and it increasingly targets the same AI-orchestration use case.
• CrewAI: CrewAI is a framework for orchestrating role-based, multi-agent LLM workflows with tool integrations. Its agent abstraction overlaps with BlobHub's OpenAI Agents and Basic Agent blocks, both targeting developers building autonomous multi-step AI applications.
• Voiceflow: Voiceflow is a collaborative platform for designing and deploying conversational AI agents and workflows. Its visual canvas with LLM, knowledge base, and tool integrations competes with BlobHub's editor for the same builder of AI-powered assistants and agents.

Broad incumbents

• Zapier: Zapier is the incumbent no-code workflow automation platform with thousands of integrations and recently added native AI actions and agents. It is a broader, more enterprise-scale incumbent to BlobHub's workflow and integration story, with a far larger installed base and go-to-market motion.
• Hugging Face: Hugging Face is the dominant hub for hosting models, datasets, and Spaces (interactive demos), with versioned model repositories and a permissioning system. Its model/dataset hosting and version control directly parallel BlobHub's Managed ONNX Blobs and Datasets Store, but at vastly larger scale.

Market position

Strengths4 records

Weaknesses4 records

Competitive moat4 records

Key risks5 records

Key highlights6 records

Customer concentration

• GitHubgithub.com/blobhubio

Revenue estimate

Valuation estimate

Number of profiles

Profiles1 record

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Blob Hub