Skip to content
LinkPress™
S01E01 AI Developer Tools Agents

The Future of AI Agents in Software Development

We dive deep into how AI agents are reshaping the way developers write, review, and ship code — from pair programming to fully autonomous pipelines.

Recording date
Duration
1:12:34
Host
Jane Smith
Guest
Alex Johnson, Maria Chen

Show Notes

6
  1. 01 00:00

    Introduction & guest welcome

    Jane introduces Alex and Maria and sets the stage for the conversation.

  2. 02 04:12

    From autocomplete to autonomous agents

    How developer assistance evolved from inline suggestions to multi-step agent loops.

  3. 03 18:40

    Tool-use and function calling in practice

    What it actually takes to let an LLM open a PR, run a test suite, and iterate.

  4. 04 33:05

    Security, sandboxing, and trust boundaries

    Designing permission models for agents that touch production systems.

  5. 05 48:30

    Impact on junior developers

    Threat, opportunity, or both? A nuanced look at the career implications.

  6. 06 01:02:15

    Predictions for the next 18 months

    Where the panel sees agentic development heading next.

Transcript

5 segments
Jane Smith 00:00

Welcome to the show. Today we're going deep on AI agents in software development with two guests who've been thinking about this from very different angles.

Alex Johnson 00:32

Thanks for having me. I've spent the last two years building agent frameworks and I think we're at a real inflection point — the gap between demo and production is finally closing.

Maria Chen 01:08

From the DX side, what's interesting is how quickly developer expectations have shifted. A year ago, autocomplete felt magical. Now teams expect their tools to actually take action.

Jane Smith 04:12

Let's start there — the shift from suggestion to action. Alex, when did you first see an agent do something genuinely useful end-to-end?

Alex Johnson 04:30

Honestly, the first time it surprised me was a small refactor task. I asked it to migrate a deprecated API across the codebase, and it didn't just generate the diff — it ran the tests, saw two failures, traced them back, and fixed the edge cases I hadn't thought of.