kademolu

3 karma · joined February 16, 2026

Cyber architecture and engineering leader building AI-native systems for security, resilience, and agent evaluation.

I work at the intersection of cybersecurity architecture, platform engineering, deterministic simulation, and AI agents. My GitHub profile (https://github.com/bluntmachetti) is my public proof-of-work space: small but serious systems, benchmarks, field notes, and experiments that explore how complex AI-enabled organisations can be tested before they touch real infrastructure, customers, or money.

submissions comments

kademolu··on Open Source Tax Engine outperforming GPT sol and Fable 5

I can't say this enough, a well tune deterministic solver always beats an LLM. In most of my experiments, the best the LLM can do is match the solvers results.

kademolu··on Agent swarms and the new model economics

Lets separate "workers" from "decision seats", you might need a lot of workers which typically don't actual require inference and might actually just be cheap solvers but actually less decision seats. So i guess it depends on what you are calling agents in your scenario.

kademolu··on You only need the frontier model for one single edit

Good stats but how does this work with models like Claude where a lot of the context or relevant information is stored in its "memory", other coding agents can't use it.

kademolu··on China’s open-weights AI strategy is winning

What a lot of people seem to miss is that you don't need the openweight models to "beat" frontier models for most agentic workloads. The harness should be doing the majority of the work, inference becomes the value add. This is where chinese models shine, i have run benchmarks where the value per dollar is always tilted to the chinese models

kademolu··on Agent swarms and the new model economics

If we take coding off the table i don't see the case for large agent swarms. More is not always better in my experience experimenting with agents