How Solomon Helps JurisTech Developers Build Faster with Agentic AI

JurisTech developers are 5x more productive with agentic AI through Solomon, our agentic development platform

JurisTech has developed Solomon, an internal agentic development platform that brings agentic AI into software development, helping our engineering teams build more effectively across complex software projects.

Solomon combines modern AI coding capabilities with JurisTech-specific engineering knowledge, development standards, and quality controls. Since its rollout, Solomon has significantly improved development productivity, with mature usage showing 5x productivity gains.

For JurisTech’s developers, Solomon provides the context that generic AI tools often lack. It understands how our teams build, test, review, and deliver software, allowing AI to become more useful in real engineering work.

Why JurisTech Built Solomon 

Before Solomon, our developers were already using coding agents such as GitHub Copilot and Claude Code. These tools delivered some productivity gains, but the impact was limited.

They worked well when developers were creating new components that required little prior knowledge of our existing systems. The limitations became clearer when teams needed to enhance or debug existing functionality.

In those situations, generic AI coding agents lacked the context needed to understand our frameworks, codebases, and project requirements. They could suggest code, but they did not always understand how JurisTech’s systems were structured or how different components worked together.

Another challenge was that these AI models were trained largely on open-source code, rather than JurisTech’s own source code. As a result, they could hallucinate functions, properties, or coding patterns that existed elsewhere, but did not belong in the JurisTech environment.

How Solomon Supports AI Software Development

We saw an opportunity to make AI more useful for our programmers by giving it the context it needed to support real JurisTech development work. Some of our strongest developers began building an internal AI platform with agentic skills, designed around our business, our codebase, and the way our teams develop software.

That work became Solomon.

Through agentic development, Solomon gives our coding agents the structure and context needed to support real development work. This includes:

• A software development methodology tailored to JurisTech’s needs.
• Component-level knowledge, including how to test and debug using our frameworks.
• Automatic translation of customer requirements into markdown files that coding agents can understand.
• Mapping of requirements to the relevant system functionality.
• Access to our Kanban boards, where software progress, issues, and bug reports are tracked.
• The ability for agents to download, track, and update issues.
• Automated code review and automated bug fixing integrated with our version control.
• Support for different coding agents, including Claude Code, GitHub Copilot, and OpenCode.
• Support across multiple environments, including MacOS, Linux, WSL, and containers.

At the heart of Solomon are two system prompt files that define the key behaviours of the agentic AI platform. These files contain Solomon’s agentic personality and development methodology, giving the platform a consistent way to guide and support our developers.

We then enhanced Solomon with more than 20 guides, organised hierarchically as optional skills for large language models (LLMs). We also added a search engine using MCP, RAG, and BM25 technologies to query key component properties and coding examples, remote browser control for testing, and 90,000 lines of Python code for automated agents and workflows.

How Solomon Helped JurisTech’s Developers Achieve 5x Productivity Gains

We first rolled out Solomon through pilot Project X.

To track progress, we used story points, a measure of work that converts requirements into stories with assigned work values.

Project X delivery trajectory after the Solomon pilot, showing the shift from generic coding-agent support to a context-aware AI development model.

Before Solomon, our developers completed 450 story points in 22 weeks using coding agents that had limited understanding of our business requirements and frameworks.

After Solomon was introduced, the team completed 1,050 story points in 15 weeks, delivering more than twice the work in 68% of the time.

The pilot delivered a 3.5x productivity gain.

As Solomon matured and developers became more effective at using it, productivity improved further. Today, our software engineering productivity is 5x greater than before Solomon.

We were also encouraged to see some of our more technical business analysts and project leads begin using Solomon to define and implement parts of the system. This showed that Solomon could support a wider group of technical contributors across our delivery process.

How Solomon Keeps AI-Assisted Software Development Anchored to Quality

As Solomon improved development speed, we also had to ensure that faster output did not lead to more bugs. To support this, we implemented Bugkilla, an automated code review and automated bug fixing engine built using LLMs.

Bugkilla combines Solomon’s agentic AI skills with JurisTech’s code quality requirements. It is integrated with our GitLab instance, allowing relevant commits to be reviewed before they move forward.

This gives our teams an additional software quality control layer within the development workflow, strengthening the way we review, validate, and improve code.

For AI-assisted software development to work at enterprise scale, speed needs to be supported by review, structure, and discipline. Solomon helps developers move faster, while Bugkilla helps keep that speed aligned with quality.

What Solomon Means for Our Customers

Solomon was built to improve how our engineering teams develop software, but its value ultimately extends to our customers.

When our teams can interpret requirements faster, navigate JurisTech frameworks more effectively, and reduce repetitive development work, they can spend more time on the implementation details that matter most. This is especially important in enterprise projects involving custom workflows, integrations, regulatory requirements, and evolving business needs.

For our customers, this supports stronger enterprise software delivery and more consistent software implementation quality. It helps JurisTech deliver reliable, scalable, and banking-grade systems with greater engineering efficiency, while keeping our development process closely aligned with real customer requirements.

How JurisTech Is Evolving Solomon for AI-Native Software Development

Solomon continues to evolve as we learn more from real development usage.

The latest enhancement is telemetry, which allows us to track how Solomon is used by our 150 developers. This gives us clearer visibility into how our teams work with the platform, where it creates the most value, and how we can continue tuning it over time.

By bringing together our codebase, development standards, frameworks, and delivery practices, Solomon gives JurisTech a stronger foundation for AI-assisted software development. As the platform matures, our focus is to keep improving productivity, quality, and engineering consistency across complex software projects.

Our goal is to keep making AI more practical, reliable, and useful for our developers, while ensuring that faster development remains grounded in disciplined engineering.

About JurisTech

JurisTech is a global lending and recovery solutions provider specialising in enterprise-class software for banks, financial institutions, insurance providers, automotive, and telecommunications companies. Founded in 1997, JurisTech supports the full lending lifecycle, from digital onboarding and origination to credit decisioning, documentation, collections, recovery, and enterprise AI adoption. Built on cloud-native, microservices, and composable architecture, its solutions help institutions modernise credit operations with greater speed, scalability, and control. JurisTech has been mentioned across multiple Gartner® reports and delivers on its motto, “360 AI Lending Tech | Fast. Proven. Secure.”