AI-assisted programming has moved far past autocomplete. Massive language fashions at the moment are able to enhancing whole codebases, coordinating long-running duties, and collaborating throughout a number of programs. As these capabilities mature, the core problem in software program growth is shifting away from writing code and towards orchestrating work, managing context, and sustaining shared understanding throughout fleets of brokers.
Steve Yegge is a software program engineer, author, and trade veteran whose essays have formed what number of builders take into consideration their work. Over the previous yr, Steve has been exploring the frontier of agentic software program growth, constructing instruments like Beads and Gasoline City to experiment with multi-agent coordination, shared reminiscence, and AI-driven software program workflows.
On this episode, Steve joins Kevin Ball to debate the evolution of AI coding from chat-based help to full agent orchestration, the technical and cognitive challenges of managing fleets of brokers, how ideas like job graphs and Git-backed ledgers change the character of labor, and what these shifts imply for software program groups, tooling, and the way forward for the trade.
Kevin Ball or KBall, is the vice chairman of engineering at Mento and an impartial coach for engineers and engineering leaders. He co-founded and served as CTO for 2 corporations, based the San Diego JavaScript meetup, and organizes the AI inaction dialogue group by Latent Area.
Please click on right here to see the transcript of this episode.
Sponsorship inquiries: [email protected]
