Skip to main content

Tasq AI

Category: AI Infrastructure

A data-centric AI platform that provides high-quality training data and human-in-the-loop (HITL) services, such as RLHF, to ensure the reliability, accuracy, and safety of enterprise-grade AI models. Tasq AI was founded in 2019. The company is led by Yoav Ziv. Based in Tel Aviv, Israel. Team size: 51-150. Total funding raised: $26.0M. Latest round: Merger (Oct 2024). Key investors include ["Fortissimo Capital","Alpha Protocol Ventures","Tal Ventures","Silicon Gardens","TLV Seed"].

Founded
2019
Headquarters
Tel Aviv, Israel
Team size
51-150
Total funding
$26.0M

Value proposition

Bridges the 'AI trust gap' by combining massive-scale human guidance with automated data orchestration, delivering high-quality, bias-free training data at a speed and scale that traditional labeling services cannot match.

Products and solutions

["NanoTasking™ Platform (Micro-tasking orchestration)","RLHF (Reinforcement Learning from Human Feedback) Services","Multimodal Data Labeling (Image, Video, Text, Audio)","AI Trust & Safety Layer (Validation and bias mitigation)","Global Localization & Cultural Adaptation (via BLEND integration)"]

Unique value

Utilizes a proprietary 'NanoTasking' approach that decomposes complex data labeling into thousands of micro-tasks, allowing for massive parallel processing and built-in cross-verification by a global crowd.

Target customer

Enterprise AI teams, LLM developers, data scientists, and global companies deploying generative AI and autonomous systems.

Industries served

["Artificial Intelligence & LLM Development","Automotive (Autonomous Driving)","Healthcare & Life Sciences","E-commerce & Retail","Finance & Legal Tech"]

Technology advantage

The merger with BLEND provides a unique advantage by combining Tasq AI's data orchestration technology with a global workforce of 25,000+ experts, enabling localized and culturally relevant AI training in over 120 languages.

How they differentiate

Tasq AI differentiates through its proprietary 'NanoTasking' technology, which breaks complex data labeling into micro-tasks for massive parallel processing. Following its merger with BLEND, it uniquely combines this orchestration with a global network of 25,000+ linguistic experts, offering localized RLHF and cultural adaptation in 120+ languages.

Main competitors

["Scale AI","Labelbox","Appen","Snorkel AI"]

Key partnerships

["NVIDIA (Inception Program member)","Major LLM providers (OpenAI, Anthropic, etc. for RLHF workflows)","Global localization and translation networks via the BLEND merger"]

Notable customers

["Meta","PayPal","Reddit","iHerb","Puma","Payoneer","NVIDIA","Xiaomi","Wix","Amazon"]

Major milestones

["Launched NanoTasking™ platform in 2020","Secured $4M Seed funding in Feb 2021","Strategic merger with BLEND (formerly One Hour Translation) in Jan 2026","Appointment of Yoav Ziv as CEO to lead the unified AI Trust entity"]

Growth metrics

Estimated combined ARR of $20M - $40M post-merger; global workforce of 25,000+ experts.

Market positioning

The 'Trust Layer' for enterprise AI, positioning itself as a high-accuracy, culturally-aware alternative to traditional mass-market data labeling services.

Geographic focus

Global, with a strong presence in North America, Europe, and Israel.

Patents and IP

Proprietary NanoTasking™ technology and AI-driven task distribution algorithms (specific patent filings are typically held as trade secrets or private IP).

About Yoav Ziv

Yoav Ziv is a seasoned technology executive with over 27 years of experience in the global software and technology sectors. Before becoming the CEO of the merged Tasq AI entity in late 2024, he served as the CEO of BLEND (formerly One Hour Translation). His career includes significant leadership roles at major tech firms, notably serving as Vice President at Checkmarx and holding various executive positions at Amdocs, where he specialized in customer success, business transformation, and operational excellence.

Official website: