Common questions buyers actually ask.

Modernize Software is the modernization practice of Semantic Designs, Inc., a software engineering firm in Austin, Texas since 1995. When you work with Modernize Software, you're hiring the same engineers, the same DMS toolkit, and the same standards that have been delivering modernization projects for three decades. The name is new. The team is not. You sign with Semantic Designs, Inc., the company behind the practice.

The reason for a dedicated practice is that the modernization market has shifted. The legacy system talent loss is accelerating, especially for COBOL, AI-led migration tools are shipping fast and producing brittle output, and compliance teams are demanding deterministic audit trails. The buyer for a COBOL to C#, COBOL to Java or other legacy language/system migration is now a distinct persona from the buyer for a general-purpose compiler tool. Modernize Software is how Semantic Designs shows up for that conversation.

Quite a lot, and the DMS Software Reengineering Toolkit powers all of it. While COBOL modernization is one of our most visible offerings, our experience spans code analysis, automated refactoring, API migrations, platform upgrades, data model transformations, and custom language-processing solutions. We help organizations make complex, cross-cutting changes across large codebases that would be difficult, costly, or risky to perform manually.

The off-the-shelf analysis tools includes the CloneDR clone-detection product, test-coverage, hardware description language analysis (VHDL and Verilog), domain-specific language compilers, formatters, obfuscators, and profilers. Modernize Software is the dedicated practice for legacy modernization, and it frequently delivers a customized solution with DMS at the core for hard-to-solve problems.

Since the first commercial engagement in the 1990s. The dedicated Modernize Software practice launched in 2025 to give the work a focused sales and service interface, but the underlying engineering is 30 years old and has run continuously on the same toolkit. The fundamental ideas were published back in 1983, as research results from Dr. Baxter's PhD thesis.

We are a small team of highly experienced software modernization specialists. Our clients work directly with senior engineers who focus on complex application transformation and modernization projects. By limiting the number of concurrent engagements, we ensure each client receives dedicated attention and access to the people doing the work.

Austin, Texas. Semantic Designs has been headquartered there since 1995. The team operates from Austin, supplemented by remote engineers across the US. For on-premise deployments at customer sites, we travel as needed, but most of the work happens in Austin.

DMS is an expert system that powers every Modernize Software project. At its core it combines symbolic AI with compiler technology to deterministically parse, analyze, pattern-match, and transform source code. In formal terms, it parses source into abstract syntax trees, builds symbol tables, records code entity definitions, runs control-flow and data-flow analysis, applies mathematically defined rewrite rules, and emits maintainable target language source code. We've been building DMS since 1995, and it remains the core technology behind our software transformation and modernization services.

A deterministic transformation is driven by explicit engineering rules; an AI-based transformation is driven by statistical prediction.

Deterministic systems apply repeatable transformations based on known source-language constructs and modernization objectives, producing consistent and auditable results. The transformation rules are built based on the mathematical principal of equivalence using the valid syntax of the legacy language as the left side and target language right side of the equation, in a form that any software engineer can recognize and validate.

AI systems generate code based on learned patterns, which can be highly useful but may produce different outputs across runs and often require additional validation. This distinction becomes particularly important in large modernization programs. When organizations need traceability, repeatability, and confidence in how business-critical logic was transformed, deterministic approaches provide a foundation that can be systematically reviewed, tested, and audited.

"In practice, we view deterministic transformation and AI as complementary technologies. Deterministic analysis provides a reliable foundation for modernization, while AI can help accelerate higher-level modernization and architectural transformation activities."

40+ languages and dialects natively, and we're adding new ones all the time. The core coverage spans COBOL and its major dialects, Java, C#, Visual Basic, Natural, Assembler, Fortran, Ada, and PL/I, along with the surrounding legacy ecosystem such as CICS, IMS, VSAM, and JCL. More importantly, our modernization capabilities are built on language analysis and transformation technology designed to adapt to new languages and dialects as needed. If you're unsure whether we support your environment, contact us and we'll assess it with you.

We regularly encounter proprietary languages, customized dialects, and unique technology environments. Unlike modernization approaches that are limited to a fixed set of supported languages, our software transformation technology was designed to be extended.

The DMS toolkit is highly flexible and configured through meta-programs built by sequencing rules, which lets us add support for new technologies without changing DMS itself. We've built front-ends for custom languages that exist nowhere else, including DowTran, Dow Chemical's proprietary 4GL. DMS handles ambiguous grammars cleanly, using GLR parsing, which is why we can take on languages that defeat ANTLR or JavaCC-based tools. Adding a new front-end usually takes a month or two for a medium-complexity language.

Yes. Modernization does not have to stop at line-for-line code translation. Major code reorganization to meet new architectural demands can be accomplished by additional sets of transformations that reshape the final code. We can transform COBOL applications into solutions aligned with modern frameworks such as Java Spring Boot or C# .NET, restructuring application components around service boundaries, APIs, and modern architectural patterns where appropriate.

On a typical project, Semantic Designs reaches 99.99% or higher fully automated code conversion. Anything that doesn't auto-convert gets flagged for engineer review. Sometimes it's more practical to address a small number of exceptional situations manually than to develop specialized transformation logic for them. The key advantage is transparency: you always know what was transformed automatically, what required manual intervention, and how the final result was produced.

We view GenAI and deterministic transformation as complementary technologies. GenAI helps accelerate system discovery, documentation, analysis, and modernization planning, while also assisting with engineering activities such as test generation and transformation development.

For the modernization itself, we rely on deterministic transformation technology to deliver consistent, traceable, and auditable results. The combination provides the speed advantages of AI together with the engineering rigor required for enterprise modernization programs.

The assessment is designed to determine modernization feasibility, risk, and likely project scope. We use automated analysis tools to examine your codebase, identify technologies and dependencies, evaluate complexity, and uncover factors that could impact a modernization effort. The findings are combined with expert engineering review to assess modernization risks, opportunities, and expected automation levels.

The result is a practical assessment report that helps you understand what can be modernized, what challenges are likely to arise, and what a modernization roadmap could look like. And if we believe a migration is not the right move at this time, we'll say so.

It depends on size, complexity and scope of modernization effort. A full portfolio migration of a multiple-million-line system often takes 9 to 24 months from start to deployment. One advantage of our highly automated approach is that customers can continue to maintain and update their legacy applications right up until the system is ready for final testing and deployment, minimizing disruption to ongoing business operations and reducing the need for lengthy code freezes.

Your legacy-system experts are among the most important contributors to a successful modernization project. They provide business and operational knowledge that cannot always be inferred from the code alone and help guide decisions throughout the modernization effort. This typically only requires a few hours of their time each week.

We work alongside your team throughout the project, incorporating their knowledge into the modernization process and preparing them to maintain and enhance the modernized application after deployment. Our goal is not to replace your team, but to equip them with a system they can successfully support for years to come.

The deliverable is not simply translated code — it is a production-ready system on your target platform. Depending on the goals of the engagement, the resulting application may reflect a straightforward language migration or a broader architectural modernization aligned with modern frameworks, service-oriented designs, APIs, workflow-based processing, and contemporary data-access patterns.

Along with the application code, we typically deliver the supporting assets needed for deployment and long-term maintenance, including database migration artifacts, test frameworks, modernization documentation, and records of any areas requiring engineering review. We also provide knowledge transfer and training to help your team take ownership of the modernized system.

The primary cost driver is the modernization target, not the conversion itself. A like-for-like migration is generally less expensive than a modernization effort that includes architectural transformation, framework adoption, database redesign, new user experiences, or cloud-native deployment models.

Because much of the transformation work is automated, organizations can often achieve substantial savings compared to manual rewrite approaches — typically 50 to 80% cost savings compared with alternative approaches. The assessment phase helps identify the desired end state, allowing the project scope and pricing to be aligned with your modernization goals.

Because predictability matters. After the assessment, we have a detailed understanding of the application, the surrounding technology ecosystem, and the desired modernization outcome. That allows us to define scope, estimate the work, and commit to a fixed-price engagement.

We are able to do that because of our automation technology, established modernization process, and decades of project experience. With a fixed-price model, the delivery risk remains primarily with us rather than being passed to the customer through an open-ended time-and-materials engagement.

Everything required to deploy, maintain, and support the modernized system. The specific deliverables depend on the project scope, but typically include the application code, supporting migration assets, testing artifacts, documentation, and knowledge transfer for your team.

There are no recurring licensing fees associated with the delivered application and no vendor lock-in. The resulting application is yours to operate and evolve.

Modernization is typically broader than code conversion. In addition to the application itself, organizations often need to plan for replacement of legacy platform services such as job schedulers, system utilities, security frameworks, etc. There is also data migration, testing and validation, deployment infrastructure, and temporary parallel operations during the transition period.

The exact mix varies by organization. Our assessment process is designed to identify these factors early so they can be incorporated into planning and budgeting before the project begins.

There is no strict minimum project size. While larger modernization programs often provide the greatest return on investment, smaller applications can also benefit from modernization depending on their business importance, technical challenges, and long-term objectives. The assessment process helps determine whether a project is a good fit and whether modernization is the most practical path forward.

Not necessarily. We can work within a variety of security models. For many engagements, our customers choose to upload their code to a secure environment established for the project. For organizations with strict security, regulatory, or classified-environment requirements, the modernization tooling can be deployed within customer-controlled infrastructure so that source code and modernization artifacts never leave their environment.

We regularly work with customer security and compliance teams to align the modernization process with established security controls and operational requirements.

You retain full ownership of both your legacy source code and the resulting translated code. You also receive perpetual, unrestricted rights to any delivered supporting project artifacts, with no royalties, recurring licensing fees, or ongoing usage charges.

Our modernization platform, transformation methodologies, and internal assets remain our intellectual property. However, the modernized application and all delivered project outputs are yours to operate, maintain, and evolve indefinitely.

Modernization projects in regulated environments require transparency and traceability. Our approach emphasizes documented, repeatable transformations, allowing organizations to understand how applications were modernized and how results were validated. Deterministic translation is easier to defend in a compliance audit than AI conversions.

We work within our customers' existing security and compliance controls, providing documentation, testing artifacts, and modernization records that can support internal reviews and audits. For environments with strict security requirements, the modernization tooling can be deployed within customer-controlled infrastructure so source code and project artifacts remain inside the organization's security boundary.

We will provide you our standard mutual NDA day one; however, if your legal team prefers your template, we'll work from that. Either way, no codebase details get discussed until the NDA is in place.

Yes. We have experience supporting modernization efforts for defense contractors and government organizations operating in highly restricted environments. When required, the modernization tooling can be deployed within customer-controlled infrastructure so source code, generated artifacts, and project data remain entirely within the customer's security boundary.

Depending on customer requirements, we can either operate the modernization process within that environment or support customer teams in running the tooling themselves after appropriate training and knowledge transfer.

Absolutely. COBOL modernization is only one part of our software transformation practice. Over the years, we have worked with a broad range of languages and platforms, including Natural/Adabas, PL/I, Assembler, JOVIAL, SabreTalk, Visual Basic, Fortran, ColdFusion, SAS, C, C++, Ada, and many others.

What makes these projects possible is not expertise in a single language, but our proprietary DMS toolkit and transformation approach that can be adapted across diverse technologies, architectures, and modernization objectives.

Yes. Software modernization is only one application of our DMS technology. We also provide standalone analysis engagements that help organizations understand, document, and manage large and complex software systems.

Common services include architecture recovery, dependency analysis, dead-code identification, business-rule extraction, impact analysis, and custom static analysis across large codebases. These projects are often performed as a precursor to modernization, but they can also provide value independently when organizations need greater visibility into critical systems.

Yes. While most organizations engage us to deliver modernization projects, others choose to license DMS directly and build their own software analysis and transformation solutions. For customers with strong engineering teams, we provide licensing, training, and advisory services to help engineering teams apply DMS to modernization challenges. Research and academic licensing options are also available.

Software modernization is only one part of what Semantic Designs does. The company also provides software analysis, code transformation, language-engineering, and developer-tooling solutions for organizations managing complex software systems.

Offerings include tools for code clone detection, programming-language tooling, HDL analysis tools for VHDL and Verilog, and custom compilers, code generators, and domain-specific language solutions. For these services and products, organizations typically engage directly with Semantic Designs.

They are often the best people to maintain and evolve the modernized system. While the technology platform changes, the business knowledge and operational expertise developed over years of supporting the application remain invaluable.

We provide training and knowledge transfer to help teams become productive on the target platform, enabling them to take ownership of the modernized application after deployment. Our goal is not to replace your team, but to equip them with a system they can confidently support and enhance going forward.

Yes. Every engagement includes a one-year guarantee covering conversion-related issues. We stand behind the modernization work we deliver.

After that period, customers may continue working with us on additional modernization or transformation initiatives, but there is no ongoing support requirement. The goal is to deliver a system your team can confidently maintain, support, and evolve on its own.

Modernization initiatives often expand beyond the first application. Once a modernization approach has been developed for a particular technology environment, future projects can frequently leverage the same analysis, transformation patterns, architectural decisions, and modernization assets.

As a result, subsequent modernization efforts are often more predictable, lower risk, and more cost-effective than the initial project. Many organizations use a phased approach to modernize portfolios over time rather than attempting a single large-scale migration.

Not every legacy system should be modernized. If an application is being replaced by a SaaS solution, if the existing system remains cost-effective and low risk to operate, or if the organization is not prepared to support the proposed target platform, modernization may not deliver sufficient value.

Our goal is not to recommend migration where it doesn't offer enough value. The assessment is designed to evaluate the business and technical case for modernization and determine whether it is the right investment. If we conclude that another path is more appropriate, we will say so.

Absolutely. Hybrid architectures are a common and often deliberate modernization strategy. Rather than forcing a full cutover, organizations frequently modernize subsystem by subsystem, moving selected business capabilities to Java or C# while the remaining legacy application continues to operate.

The modernized and legacy components are integrated through well-defined interfaces, allowing both environments to coexist and evolve together. This approach reduces risk, minimizes business disruption, and enables modernization to proceed incrementally based on business priorities.

Any code that cannot be translated with complete confidence is flagged for engineer review rather than being silently converted. This is one of the most important differences between deterministic transformation and AI-generated translation. An AI tool may produce output that appears correct even when it does not fully understand the source code. DMS does not.

The vast majority of code is converted automatically. The small remainder typically involves unusual language constructs, library calls that have no direct equivalent in the target environment, or particularly complex business logic where human review is warranted.

That's common in legacy environments. When automated tests are limited or unavailable, we use a combination of automated analysis, regression-validation techniques, and test-generation tools to establish confidence in the modernized system.

Where appropriate, AI-assisted and DMS-based tooling can help generate test cases and improve coverage of important business functionality. The exact testing strategy depends on the application's complexity, risk profile, and business requirements.

The objective is not just to change languages — it is to create a system your Java or C# developers can successfully maintain. Modernization can go beyond preserving legacy structures and instead align the application with contemporary development practices, frameworks, and architectural patterns.

Depending on the scope of the engagement, that may include improved code organization, stronger abstractions, service-oriented designs, and framework-native implementations that are easier to support and extend over time.