Skip to content
← Tools / Skills
ToolKodesage/10 trustVerified

Kodesage

On-premise AI platform that documents, analyzes, and modernizes complex legacy enterprise software.

Visit official site →

About

Kodesage is a startup founded in 2024 by Gergely Dombi, Miklos Szurdi, and Gyorgy Szilagyi, building an on-premise AI platform for legacy software modernization. In June 2026 it raised $6.6 million in seed funding led by VentureFriends, with participation from Portfolion and angels including Christian Szegedy (xAI co-founder) and Mario Goetze (all per press coverage).

The platform extracts information from source code and documentation to build a continuously updated knowledge layer over mission-critical applications, aimed at regulated industries like banking, insurance, energy, and the public sector. It analyzes both modern and legacy stacks (COBOL, Oracle Forms, PL/SQL, PowerBuilder, RPG) and automates codebase discovery, documentation generation, context-aware code conversion, test creation, and AI-assisted production support.

The honest picture: the defining trait is deployment entirely inside customer-controlled environments (on-premise, VPC, or air-gapped), which is what makes AI tooling palatable to regulated enterprises. The company talks about self-healing enterprise applications as a direction, which is roadmap language rather than a shipped capability. Funding and traction figures are press-reported.

What it can do

  • Living knowledge layer extracted from code and docs
  • Codebase discovery and documentation generation
  • Context-aware legacy code conversion
  • Test generation and AI-assisted production support

Setup

Enterprise engagement; platform deploys inside the customer's own environment.

Prerequisites

An enterprise legacy codebase and infrastructure for on-premise or VPC deployment.

Use cases

legacy-modernizationcode-analysisdocumentationenterprise

Works with

Trust score — 8/10

/10 trustVerified

Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 114; Trending score 78/100 from researcher assessment.

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.