AI Coding & Productivity - AI is considered a significant technological wave with the potential to create substantial economic impact and productivity gains [3] - The development of effective agentic solutions, particularly in software and coding, has become visible, with widespread adoption of AI assistance in coding since the introduction of GitHub Copilot [4][5] - The industry anticipates that the paradigm of AI assistance will extend to other areas of software development and various industries beyond coding [5] Challenges & Solutions - Generating more AI code without addressing subsequent steps can be a liability, increasing incidents and making code maintenance harder [12][13] - The industry believes the solution lies in applying AI to monitor, maintain, and troubleshoot AI-generated code to improve overall velocity [14][15] - Resolve AI focuses on building AI solutions that prioritize trust for software engineers, allowing AI to investigate and propose solutions, with human oversight before full automation [17][18] Future of AI in Software Engineering - The industry predicts that within a year, AI will become the primary driver of software, with humans overseeing at a higher level, and within two to three years, AI will make most decisions [20] - The industry emphasizes the importance of deep agentic applications that understand the domain and customer context, requiring innovation in models to handle more data and longer task horizons [27] - Multi-agent systems with various layers of guardrails, checks, and validations are crucial for reliable AI performance, with an orchestrator agent managing other agents [31][32][33] AI Model Specialization - The most capable and expensive AI models are typically used at the top level for reasoning and planning, while specialized or open-source models can handle underlying tasks [34][36] - The industry anticipates that domain-specific large models will emerge for areas like software and customer service due to their significant economic impact [36] Adoption & Cultural Impact - While engineers are early adopters of AI, there is some resistance to change and concerns about job security [38][39] - The industry believes the goal is to produce technology faster, benefiting the world, and engineers will operate at a higher level of abstraction, with AI handling low-level tasks [40][44]