K2view

Primary category: synthetic-masked-data.

One-liner — An Israeli enterprise data-platform company whose entity-based architecture (one “micro-database” per customer, account or order, assembled from every source system) powers test data management, data masking and synthetic data, as well as data integration and, more recently, AI-agent data access.

What it does — K2view sells one platform (Data Product Platform, earlier Data Fabric) used for several jobs. The test-data job is the relevant one here:

• Test Data Management (K2 TDM): subset test data “using familiar business terms” (e.g. all accounts of customers in a given state with a given balance). Masking happens in flight from a library of masking functions, with referential integrity preserved across systems. Plus reservation (stop testers overwriting each other’s data) and snapshot/rollback.
• Synthetic data: “Generate synthetic data using AI, rules, or entity cloning”. Cloning means replicating a masked entity many times with varied values for volume testing.
• Sensitive data discovery: discovers and catalogs PII across sources to drive masking rules. This exists to feed masking. It is not marketed as a posture-management (DSPM) product, so the page is not cross-listed to dspm.
• Outside this category, the same platform does operational data integration (customer-360 data products), and since 2024–2025 sells “AI-ready data” for RAG and agents, including an auto-generated MCP server per data product.

Naming / provenance — Founded 2009 by Achi Rotem and Rafi Cohen (ex-Amdocs/telecom), HQ Yokneam, Israel, with a US office in Palo Alto. Heavy telco heritage (Verizon, AT&T, Vodafone, Cellcom, Pelephone named). Product names drift between “Data Fabric”, “Data Product Platform”, “K2 TDM”, “AI Data Fusion” and “AI Context”. Survey respondents will just say “K2view”.

Ownership & viability — independent. The 2020 $28M round was led by Forestay Capital with Genesis Partners (Wikipedia, not independently checked). The latest confirmed financing is $15M growth capital from Trinity Capital, announced 2025-06-23. Trinity is a listed venture-debt/growth lender, so this is likely debt-like capital rather than a priced equity round (terms not disclosed). Leadership: Ronen Schwartz was CEO at the 2025 raise. Wikipedia reports Steve Kostyshen as CEO in 2026 with F. Thaddeus Arroyo as chairman. No primary source was found; treat as unverified. At 17 years old and modestly capitalised, K2view is a plausible acquisition target for a data-integration or DevOps vendor (compare Perforce/Delphix). On-prem customers would keep running licensed software.

Positioning & differentiators — challenger in synthetic-masked-data: a credible enterprise TDM alternative to the legacy suites and to tonic-ai, but smaller and less visible to developer buyers.

• Entity-based versus database-based. tonic-ai and Delphix work database by database. K2view assembles a business entity across all systems first, then masks and subsets it. That keeps a customer consistent across, say, CRM, OMS and fund-admin extracts, the hard part in large estates.
• Versus legacy TDM (Informatica, Broadcom TDM, Delphix/Perforce): K2view sells itself as the modern replacement. Its insurance case study is titled as replacing in-house TDM tools, and Bloor Research profiles its TDM.
• Versus gretel (NVIDIA Safe Synthesizer): K2view’s synthetic data is aimed at test volumes and coverage, not at statistically faithful ML training sets with differential privacy.
• Breadth cuts both ways. It is a whole data platform. Buying it for TDM alone is heavier than buying a point tool.

Who should choose them / anti-fit — Fits a large, multi-system shop (bank-affiliated asset manager, multi-strategy fund with a big internal engineering org, or a firm already using K2view for data integration) that needs consistent masked test data across many systems of record, including legacy and mainframe sources. Anti-fit: a typical hedge fund with a handful of databases and a small dev team (Tonic or open-source masking is lighter); teams wanting self-serve, developer-priced tooling; and anyone mainly after synthetic data for ML training.

Known weaknesses / gotchas —

• Platform weight. Implementation is a project, with entity modelling up front. Expect professional services.
• No public pricing. Enterprise sales motion only.
• Marketing sprawl. Recent messaging leans heavily on agentic-AI “context” products. Confirm what you are actually buying and that TDM remains a funded line.
• Customer evidence is mostly anonymised (“Fortune 500 Bank”, “Fortune 500 insurer”), with vendor-reported metrics.
• Leadership-change report is unverified. Check who runs the company before a multi-year commitment.

Deployment & data handling — “Deploy across on-prem, cloud, and hybrid architectures” (vendor). On-prem/customer-cloud deployments keep production data in the customer’s environment. Whether a K2view-hosted SaaS processes customer production data, and where, is unverified (data_leaves_tenancy: unverified). Masking happens in flight during extraction, so unmasked data should never land in the test environment. Batch/on-demand provisioning, not inline.

Integrations & partnerships — Connectors to relational databases, SaaS and enterprise apps. K2view blogs cite SAP, Salesforce and SQL Server; the full list was not reviewed. MCP: the platform auto-generates an MCP server per data product, with governance and anonymization applied, giving agents masked, governed access to entity data. That is notable but outside this category’s core job. Gartner named K2view a Visionary in the 2025 Magic Quadrant for Data Integration Tools (not a TDM rating). Its TDM page lists Gartner’s synthetic-data market guide as a resource.

Compliance & FS traction — Certifications not verified. FS traction claimed: Wikipedia lists Charles Schwab and KeyBank as customers. K2view publishes an unnamed “Fortune 500 Bank” TDM case study (a brokerage/wealth firm with multi-trillion client assets) claiming test-data provisioning staff cut from 50 to 10 FTEs. Vendor figures, customer not named. No hedge-fund or asset-manager customer found.

Commercial — Not public. Enterprise licence (likely subscription, priced by deployment scope; unverified).

Open questions

• Who is CEO in 2026 (Wikipedia’s Kostyshen/Arroyo report needs a primary source)?
• Total funding; nature of the Trinity Capital facility (debt vs equity).
• SOC 2 / ISO 27001; whether a K2view-hosted SaaS exists for TDM and where it processes data.
• Pricing model and typical entry cost for a mid-sized firm.
• Named FS references (is the “Fortune 500 Bank” Schwab?).

Sources

• K2view Test Data Management — fetched 2026-10-26 — supports: TDM capabilities, deployment, Gartner DI MQ; confidence: medium (vendor marketing)
• K2view secures $15M (K2view news) — fetched 2026-10-26 — supports: 2025-06-23 Trinity Capital financing, CEO at the time; confidence: high
• K2view (Wikipedia) — fetched 2026-10-26 — supports: founding, HQ, 2020 round, customer names, product history; 2026 CEO change (unverified); confidence: low
• TDM case study: Fortune 500 bank (K2view) — via search 2026-10-26 — supports: anonymised bank case study; confidence: low (vendor)
• K2view MCP integration — via search 2026-10-26 — supports: MCP server per data product; confidence: medium (vendor)
• Cached: raw/sources/2026-10-26--k2view--tdm-funding-mcp.md

History

• [2026-10-26] Page created via wiki-create and researched the same day, as a seed vendor of the new synthetic-masked-data category. Not cross-listed to dspm: its sensitive-data discovery feeds masking and is not sold as posture management.