Version 1

version1.com

Version 1 is a technology delivery provider rather than a research advisory business or solely a software vendor. Buyers can commission technology assessments, implementation projects and ongoing support. Its AI practice explicitly covers strategy, production applications, governance, security assessment and operations. Separate offerings address cloud infrastructure, enterprise data, legacy applications and Oracle ERP. Procurement scope should distinguish advisory deliverables from engineering and post-launch responsibility. ^[1] ^[2] ^[3] ^[4] ^[5] ^[6] ^[7]

Services and delivery

AI strategy and use-case assessment

Advisory. Assesses AI maturity and capability gaps, prioritises use cases and develops business cases and roadmaps with ROI targets. This is advisory work, distinct from prototype or production delivery. ^[1]

AI application prototyping and production engineering

Implementation. Designs, builds and assures production AI services for knowledge retrieval, decision support, agent-driven workflows and customer experiences. Deliverables include enterprise knowledge assistants respecting existing access controls, multimodal applications and progression from proof of concept to production. ^[2]

AI governance advisory

Advisory. Helps organisations address AI regulatory requirements through governance policies, accountability arrangements, use-case risk classification and gap-remediation plans. ^[3]

AI security assessment

Advisory. Assesses AI-specific exposure through shadow-AI discovery, model vulnerability testing, red teaming and sensitive-data exposure assessment. ^[3]

Production AI support and optimisation

Managed services. Provides ongoing support for production AI solutions and models, addressing accuracy, relevance and performance. The offering also covers visibility and optimisation of AI tooling, token usage and model consumption. ^[3]

Cloud migration and platform implementation

Implementation. Executes migrations using wave planning and dependency-aware sequencing. Scope includes platform design and build, identity and networking foundations, security, DevSecOps pipelines, legacy migrations and database migration across AWS, Azure and Oracle cloud environments. ^[4]

Governed enterprise data-platform implementation

Implementation. Builds platforms, ingestion and transformation pipelines, data-quality controls, lineage, cataloguing and stewardship frameworks. Delivery includes DevOps pipelines, documentation, runbooks and knowledge transfer. Named platforms include Microsoft Fabric, Azure Synapse, Databricks, Snowflake, AWS and Oracle. ^[5]

Data-platform managed services

Managed services. Supports Azure, AWS and Snowflake environments, including pipeline reliability, governance, integration and business-intelligence support. Available models include time-and-materials work, proactive managed services and reactive drawdown support. ^[8]

Oracle Cloud ERP roadmap advisory

Advisory. Provides roadmap advisory for Oracle Cloud ERP, separately from configuration, integration and migration delivery. ^[7]

Oracle Cloud ERP implementation and integration

Implementation. Configures ERP modules, integrates systems, migrates legacy data and performs reconciliation and testing. The offering is implementation work around Oracle software, not merely licence supply. ^[7]

Oracle Cloud ERP managed services

Managed services. Provides ongoing Oracle Cloud ERP support, including quarterly release management and enhancements. ^[7]

Legacy application modernisation

Implementation. Offers application transformation, automated code modernisation and embedded engineering squads. This service is distinct from the firm’s named code-analysis accelerator. ^[6]

Decipher code-analysis solution

Software product. A generative-AI static code-analysis solution and delivery accelerator for application documentation, business-rule and integration-point extraction, and documentation assistance. The datasheet identifies Azure OpenAI, AWS Bedrock and graph-database technology, but does not establish standalone licensing terms. ^[9]

Company details

Engagement and delivery

AI engagements can begin with a bounded Co-Creation Kickstarter covering readiness, discovery and a working prototype. Subsequent options include delivery sprints and multidisciplinary squads that design, build and run solutions. Production services also offer embedded specialists and Forward Deployed Engineers. ^[1] ^[2]

Track record

For insurer Foundation Risk Partners, Version 1 reports building AI-assisted policy-document analysis with human validation. It reports reducing gap analysis from 40 hours to approximately four hours and wedge analysis from four hours to around 15 minutes. These are customer-specific reported outcomes, not general commitments. ^[10]

For The National Museum of Computing, Version 1 developed TNMOC Mate using GPT-4 to adapt exhibit information for different ages, languages and neurodiverse profiles. Its SpiderRock case instead documents a bounded assessment: five Perl files were converted using ChatGPT and validated against actual outputs, with reported conversion-time reductions of up to 87%. ^[11] ^[12]

The Hays case documents discovery, licensing assessment, migration planning and execution across AWS, Azure and Oracle Cloud Infrastructure, followed by PeopleSoft and cloud managed services. Version 1 reports no downtime or data loss for that migration. ^[13]

For Enterprise Ireland, Version 1 documents Oracle Fusion Cloud ERP implementation covering General Ledger, Accounts Payable and Receivable, and Cash Management, alongside bespoke tax and mileage-expense solutions and continuing managed services. ^[14]

Scope and limitations

The supplied evidence confirms generative-AI consulting and delivery, but does not establish every form of custom predictive-model development or training.
Data-platform implementation and operations are documented separately from AI delivery; the data managed-services page alone does not substantiate an AI managed service.
The SpiderRock assessment covered five files and does not establish conversion or deployment of the entire legacy codebase.
Foundation Risk Partners’ financial figures were inconsistent within the supplied research and are not reproduced here.
Several broader practice pages were not fully retrievable during the supplied research. Conclusions rely on the specialist service pages and case studies recorded in the notes.
Standalone Decipher licensing, delivery locations, contractual service levels and commercial terms are not established by the supplied evidence.

Questions to ask

Which assessment, prototype, production and operational deliverables will Version 1 own, and what are the acceptance criteria?
How will AI quality, model evaluation, human review and security testing be specified and maintained?
What data-residency, access-control and third-party model constraints apply to the proposed architecture?
Who owns delivered code, documentation and accelerator-dependent components, and what additional licences are required?
What support coverage, incident responsibilities, service levels and exit arrangements apply after go-live?
Can Version 1 provide references matching the required platform, migration complexity and, where relevant, custom predictive-ML scope?

Overview