Growing & Thriving in a Multi-model World • Alberto Brandolini • GOTO 2025

2026-06-17 · Source: GOTO Conferences · Field: Technology & Digital — Software Development & Engineering, Corporate Strategy & Leadership, Project & Product Management · Depth: Advanced, extended

Summary

Alberto Brandolini's GOTO 2025 talk, "Growing & Thriving in a Multi-model World," addresses the inherent complexity in software development when relying on single, monolithic models. He illustrates how projects often start simple but inevitably grow into "holy messes" due to ignored boundaries, leading to increased dependencies, delayed refactoring, and escalating coordination costs. Brandolini advocates for a Domain-Driven Design approach, emphasizing the use of bounded contexts—models with single purposes, well-defined boundaries, and autonomy. He shares insights from a 10-year in-house project, highlighting the flat cost of evolution achieved through disciplined modeling, visual context mapping, and structured architectural discussions. The talk also explores challenges in scaling to market fit and multi-country operations, stressing the need for early, purpose-driven model splitting and the development of "uber domain experts" to navigate diverse market regulations and cultures.

Key takeaway

For software architects and engineering leaders grappling with growing system complexity, you should proactively implement Domain-Driven Design's bounded contexts and visual context mapping. This approach, focusing on purpose-driven model splitting, reduces long-term evolution costs and prevents monolithic "holy messes." Prioritize early, cheap decision-making by involving fewer people and documenting architectural discussions with ADRs. Don't fear starting small; even minor refactoring driven by pain or fear can initiate a beneficial cleanup process.

Key insights

Software complexity is best managed by embracing multiple, purpose-driven models with clear boundaries from the outset.

Principles

• Models should have a single, sharpened purpose.
• Don't change working design unless purpose changes.
• Decisions are cheaper with fewer people involved.

Method

Employ context maps for visual decision-making, framing architectural discussions with problem visualization, alternative solutions (at least three), and capturing outcomes with Architecture Decision Records (ADRs).

In practice

• Use context maps to identify new feature placement.
• Reject terms not suitable for a model's purpose.
• Explore LLMs to get plausible foreign market expertise.

Topics

• Domain-Driven Design
• Bounded Contexts
• Software Architecture
• Context Mapping
• Model Splitting
• Technical Debt
• Organizational Scaling

Best for: AI Architect, VP of Engineering/Data, Director of AI/ML, Software Engineer, CTO, Consultant

Related on AIssential

• The Architecture Shift Behind Reliable Enterprise AI - with Ravi Marwaha of Arango — AI scaling failures stem from fragmented data context, not model limitations, requiring a re-architecture for agentic systems.
• Scaling beyond one: How Airbnb evolved its data architecture for a multi-product world — Balancing centralized data consistency with domain-specific flexibility is crucial for scalable multi-product data architectures.
• How to Navigate the Shift from Prompt-Based Tools to Workflow-Driven AI — The proliferation of specialized AI tools creates workflow friction; unified platforms offer a solution by integrating models and context.
• Article: Architectural Change Cases: A Practical Tool for Evolutionary Architectures — Architectural change cases anticipate future system evolution and decay, guiding proactive design decisions.
• Why One AI Model Will Never Be Enough — No single AI model can be "best" across all tasks; specialization is key for optimal performance.
• Supporting Design Decisions in Rule-Based Model Transformations — Explicitly modeling design decisions in model transformations improves flexibility, reusability, and traceability.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by GOTO Conferences.