Client Overview
The client is one of Japan’s leading providers of desktop accounting software, and thousands of Japanese accounting firms and small businesses rely on its flagship software every day. It is a dominant platform in the region, built and refined over a twenty-year history to serve as the absolute backbone of their customers’ daily financial operations.
But two decades of continuous evolution come with a catch. While the application remained highly reliable and packed with features, years of layered enhancements left a footprint of tangled legacy code, aging database technology, and heavy maintenance overhead. As market expectations shifted, leadership knew they needed to upgrade, but they had to do so without disrupting their users’ daily workflows.
Rather than gambling heavily on an immediate, massive overhaul, the client opted for a targeted pilot program to see whether AI-assisted engineering could actually handle their complex tech debt, protect application performance, and give them the concrete proof needed to greenlight an enterprise-wide rollout.
Customer Challenge
Key challenges
Legacy application architecture
The software relied on outdated Shift_JIS text encoding, JET 4.0 databases, and deeply intertwined desktop components. This combination bottlenecked scalability and turned routine maintenance into a major chore.
High modernization risk
Rip-and-replace strategies for a core accounting engine carry massive operational risk. There was simply zero room for error or downtime.
Performance expectations
Accountants move fast. The upgraded system absolutely had to match the lightning-fast responsiveness, speed, and keyboard-driven navigation that professionals have relied on for decades.
Business continuity
Existing accounting workflows, seamless offline functionality, and the familiar user interface couldn’t be sacrificed during the transition.
What was at stake?
Jumping straight into a massive transformation without proof meant risking major capital on a guess. They had no real-world evidence that modern frameworks, cloud architectures, or AI tools could meet their technical and business requirements.
The company needed a safe testing ground. They required a pilot that could stress-test their technology choices, benchmark real performance, calculate actual AI acceleration rates, and layout a realistic blueprint for the rest of the software.
Ness Solution
Ness teamed up with the client for a focused five-month pilot to tackle their most critical feature, the General Journal module. Using our ATONIS AI acceleration platform, we rolled up our sleeves to modernize legacy code, migrate the core database, and run side-by-side tests of cloud microservices against offline-ready architectures before mapping out the next steps.
This wasn’t just a theoretical exercise; it combined hands-on AI engineering, database overhauls, and rigorous architectural testing to strip away deployment risks while building a highly scalable foundation.
Key initiatives
AI-assisted modernization
We used ATONIS AI to automate the tedious Unicode conversion process, successfully updating roughly 150,000 lines of legacy General Ledger code.
Database modernization
We migrated the aging JET 4.0 database engine over to PostgreSQL, instantly improving the system’s scalability and future maintainability.
Application modernization
Using .NET MAUI, we rebuilt a native Windows application that felt modern under the hood but kept the exact user experience and workflows accountants expected.
Architecture validation
We built and compared cloud-based microservices against an offline-first, local PostgreSQL setup to find the perfect fit for their users.
Performance and business validation
We rigorously measured system speed, UX friction, AI automation efficiency, and overall ROI to give stakeholders hard data for their final decisions.
Engagement model
The Ness team worked side by side with the client’s internal engineers and business units throughout the five-month sprint. By blending our expertise in AI engineering, Windows development, and cloud architecture with their deep domain knowledge, we created a phased rhythm. This kept everyone aligned and allowed us to prove concepts without risking business continuity.
What Changed?
The client walked away with a fully verified, risk-tested roadmap for their entire platform.
Business Outcomes
Strategic impact
The client successfully proved an AI-driven modernization strategy without exposing the business to the chaos of a blind, full-scale launch. By isolating a single, high-stakes module, testing multiple ways to deploy it, and measuring the exact technical outcomes, they bought themselves total confidence before spending a dollar on large-scale investments.
With a freshly minted application foundation, a modern PostgreSQL database structure, and validated AI engineering processes in place, the client is sitting in a perfect position. They can now aggressively scale modernization across their entire software suite while slashing technical debt, avoiding implementation traps, and driving long-term product innovation.
Let’s Engineer What’s Next. Together.
Partner with us to build intelligent solutions faster and smarter — we’re ready when you are.
Our "Contact Us" webform relies on a tracking cookie. Your current cookie preferences do not permit these cookies. To contact us through our "Contact Us" webform, please ["Allow All"] cookies in Manage Cookie Settings option in our Cookie policy. Alternatively, you can email us directly at [email protected].
