Multi-Agentic Software Development Is a Distributed Systems Problem
by tie-in on 4/14/2026, 5:32:48 AM
https://kirancodes.me/posts/log-distributed-llms.html
Comments
by: xer
The fundamental assumptions of distributed systems is having multiple machines that fail independently, communicate over unreliable networks and have no shared clock has the consequence of needing to solve consensus, byzantine faults, ordering, consistency vs. availability and exactly-once delivery.<p>However, AI agents don't share these problems in the classical sense. Building agents is about context attention, relevance, and information density inside a single ordered buffer. The distributed part is creating an orchestrator that manages these things. At noetive.io we currently work on the context relevance part with our contextual broker Semantik.
4/14/2026, 12:02:32 PM
by: yangshi07
After you point this out, it is obviously right!
4/14/2026, 12:37:05 PM
by: 21asdffdsa12
To be honest humans often have no overview over a application either. We navigate up and down the namespace, building the "overview" as we go. Nothing i see what prevents an agent from moving up and down that namespace, writing his assumptions into the codebase and requesting feedback from other agents working on different sets of the file.
4/14/2026, 11:39:05 AM
by: SamLeBarbare
Conway’s law still applies.<p>Good architecture, actor models, and collaboration patterns do not emerge magically from “more agents”.<p>Maybe what’s missing is the architect’s role.
4/14/2026, 9:11:48 AM
by: falcor84
The thing that TFA doesn't seem to go into is that these mathematical results apply to human agents in exactly the same way as they do to AI agents, and nevertheless we have massive codebases like Linux. If people can figure out how to do it, then there's no math that can help you prove that AIs can't.
4/14/2026, 6:43:26 AM
by: jbergqvist
Doesn't this whole argument fall apart if we consider iteration over time? Sure, the initial implementation might be uncoordinated, but once the subagents have implemented it, what stops the main agent from reviewing the code and sorting out any inconsistencies, ultimately arriving at a solution faster than it could if it wrote it by itself?
4/14/2026, 6:55:49 AM
by: lifeisstillgood
It’s not a solution but it’s why humans have developed the obvious approach of “build one thing, then everyone can see that one thing and agree what needs to happen next” (ie the space of P solutions is reduced by creating one thing and then the next set of choices is reduced by the original Choice.<p>This might be obvious to everyone but it’s a nice way to me to view it (sort of restating the non-waterfall (agile?) approach to specification discovery)<p>Ie waterfall design without coding is too under specified, hence the agile waterfall of using code iteratively to find an exact specification
4/14/2026, 7:42:14 AM
by:
4/14/2026, 9:50:00 AM
by: vedant_awasthi
Makes sense. Coordination between multiple agents feels like the real challenge rather than just building them.
4/14/2026, 10:12:21 AM
by: cooloo
Well it starts with agree list. I don't agree next gen models will be smarter. I would argue no real improvement in models in last couple of years just improvement in stability and tools (agentic ones) around it.
4/14/2026, 10:24:42 AM
by: enesz
[dead]
4/14/2026, 7:52:10 AM