Posts by Flamehaven

@Flamehaven

Yun Kwansub

Scientific AI Systems Architect | BioAI & Physics-ML | Deploying Regulated AI Mo...
South Korea flamehaven.space Joined March 2026
4.4k Points118 Badges28 Connections64 Followers32 Following

Posts by Flamehaven

Flamehaven in Articles 13 min read
Ask how much of the code shipping today is AI-authored and even the headline number turns out to be harder to define than it looks. Veracode’s 2026 GenAI Code Security Report cites research from DX putting the figure at 51.9% 1. But the underlying D...
post-cover-24582
Flamehaven in Articles 20 min read
> 💡TL;DR > > > ▪️GraphRAG helps most when questions require multi-document relationships, temporal reasoning, or corpus-level synthesis; it does not consistently beat standard RAG on direct factual retrieval. > ️️▪️Automatically constructed graphs ...
post-cover-23790
Flamehaven in Articles 12 min read
The Slop Gap: Why Rising Model Capability Hasn't Solved Code Quality And what happens when you turn “slop” into a falsifiable engineering hypothesis The gap nobody's benchmark shows !2https://coderlegion.com/?qa=blob&qablobid=7984006102791652378 ...
post-cover-23618
Flamehaven in Articles 13 min read
Translating Parcae's Stability Questions into an Agent-Level Control Loop 💡TL;DR — Across three fresh seeds on the same 42-fact development benchmark, an offline-informed two-sample fallback cap moved four or five facts per seed from fail to pass ...
post-cover-23121
Flamehaven in Articles 15 min read
The Illusion of the Green Light !https://coderlegion.com/?qa=blob&qablobid=9142298931602403537 Not long ago, porting code from one language to another was a job for human hands. You read a line, doubted it, moved it, read it again, stopped when so...
post-cover-22143
Flamehaven in Articles 8 min read
In short: PYRHELIX is an offline dual-control release gate for sensitive BIO/PII artifacts. It does not replace perimeter security, compliance review, or clinical judgment. It closes one specific gap: no single actor, key, or approval should be able ...
post-cover-21756
Flamehaven in Articles 23 min read
The Comment A few weeks ago I posted something I had put real care into. A public verification ledger, dozens of records, a session-start contract that checks an AI maintainer's working state before it touches an archive. The kind of post where the...
post-cover-21606
Flamehaven in Articles 15 min read
When a Health-AI Release Becomes a Governance Surface !2https://coderlegion.com/?qa=blob&qablobid=6341999130821584442 Biotech and health-AI didn't slow down in 2026. By June, the Flamehaven team's tracking of disclosed biotechnology equity rounds...
post-cover-21414
Flamehaven in Articles 11 min read
!https://coderlegion.com/?qa=blob&qablobid=14914080016213615833 Part 9 of the MICA series — Read from the beginning1 The Question Nobody's Asking !2https://coderlegion.com/?qa=blob&qablobid=8524171406895368086 The advice is consistent enough to ...
post-cover-21353
Flamehaven in Articles 12 min read
The shape of the work right now !1https://coderlegion.com/?qa=blob&qablobid=14808488737358335165 Zenodo views: 1. Downloads: 0. No external citations to the ledger to date. The code still runs. That is the uncomfortable shape of this work right ...
post-cover-21189
Flamehaven in Articles 11 min read
!titlehttps://coderlegion.com/?qa=blob&qablobid=7182322320887322452 Why there is a second post About six months ago we built and published the original QSOT artifact. The name — Quantum State Over Time — promised a deeper link between quantum evol...
post-cover-21011
Flamehaven in Articles 11 min read
💡 Note: This is not an argument to dismiss the Nature Medicine paper. It is an argument for stronger validation infrastructure around medical AI benchmarks before practice-shaping claims become settled wisdom. Two Critics, Two Reasonable Conclusion...
post-cover-20720
Flamehaven in Articles 10 min read
Not just a new framework, but a clearer answer to what the score means, why the report exists, and how the artifact should be read. The real change in v1.8.0 through v1.8.4 was not that STEM BIO-AI cited one more framework. The real change was th...
post-cover-20388
Flamehaven in Articles 6 min read
!coverhttps://coderlegion.com/?qa=blob&qablobid=1036492842589157619 An AI agent fixed a release for me. That sentence sounds cleaner than the session felt. What actually happened: the agent audited the documentation, found a public-output problem...
post-cover-20322
Flamehaven in Articles 18 min read
We Made a High-Formality, Fake Physics Slop Artifact QSOT Quantum State Over Time Compiler - A Post-Mortem on AI-Native Paper Implementation Gone Wrong > "The most dangerous form of scientific fraud is not the one that looks wrong. It is the one ...
post-cover-20084
Flamehaven in Articles 12 min read
On where craftsmanship went, why verification gaps appear in its absence, and the one practice AI cannot automate for you. From Writing Code to Judging Code There is a sentence that has been circulating through developer circles since 2026, attri...
post-cover-19843
Flamehaven in Articles 8 min read
!coverhttps://coderlegion.com/?qa=blob&qablobid=443347642390904533 For a long time, the hardest part of software development was writing code. That is no longer true. As AI-assisted coding and agent-driven workflows become mainstream, the cost of g...
post-cover-19775
Flamehaven in Articles 10 min read
When the Memory Gate Met a Real Archive This article is the practical side of the MICA series. MICA means Memory Invocation and Context Archive. In this workflow, it is a small package loaded at the start of a maintenance session so the active rule...
Flamehaven in Articles 7 min read
Prologue: The Mirage & The Pivot In the summer of 2024, our team began with an ambition that felt almost impossible: to use frontier AI for drug discovery, theoretical physics, and problems that seemed unreachable. We imagined systems that could re...
post-cover-19525
Flamehaven in Articles 9 min read
!1https://coderlegion.com/?qa=blob&qablobid=11961288196514487540 In 2024 and 2025, YouTube updated its monetization policies to explicitly exclude "repetitious" and "mass-produced" content from the YouTube Partner Program 1https://support.google.com...
post-cover-19170
chevron_left

Latest Jobs

View all jobs →