▶Lorentzon consistently argues that AI is fundamentally changing the nature of software development, citing that over 50% of Lagora's code is AI-generated and that the bottleneck has shifted from writing code to reviewing it and defining products.Jun 2026

▶He repeatedly emphasizes that the role of the software engineer is evolving away from implementation towards higher-level abstraction, focusing on systems design, architecture, and managing AI agents.Jun 2026

▶Lorentzon advocates for a pragmatic, multi-model approach to AI implementation, stating Lagora uses approximately 10 different models and frequently switches between providers like OpenAI and Anthropic based on performance for specific tasks.Jun 2026

▶He views AI as a core driver of internal business operations, enabling the creation of custom internal systems for HR and payroll through a process called 'vibe coding' and improving processes like incident response.Jun 2026

▶Lorentzon champions massive AI-driven productivity gains for engineers but also projects a more than threefold growth in his engineering headcount by 2027 (from 80 to 270), suggesting that increased output demand will significantly outpace individual efficiency improvements.Jun 2026

▶He notes that AI makes prototyping by non-engineers (like Product Managers) highly efficient, yet argues it's inefficient for PMs at a company like Lagora to code due to the high opportunity cost of not performing core product work.Jun 2026

▶Lorentzon expresses concern about a potential duopoly or monopoly in foundation models and hopes for more open-source competition, while simultaneously acknowledging his company's heavy reliance on the top market players (models from Anthropic and OpenAI) for critical tasks.Jun 2026

▶He believes AI makes threat actors 'extremely efficient' and creates a new generation of security threats, raising the question of whether defensive AI capabilities are keeping pace with offensive ones.Jun 2026