AI/ML●●●Banger
I built a 2nd-order PyTorch optimizer for LLMs that runs on 16GB GPUs
Runs Shampoo-quality second-order optimization on a 16GB T4 where others OOM immediately.
WizardryBig Brain
dnosoz
241mo ago
Back to browse
GitHub Repository
A PostgreSQL extension that can enforce optimal column alignment to minimize row padding waste.
96 starsPLpgSQL
Pg_column_Tetris – a pg extension for optimal column alignment
by rogerw·Apr 30, 2026·2 points·0 comments
AI Analysis
●●SolidBig BrainSolve My Problem
Blocks suboptimal CREATE TABLE inside Postgres when SQLFluff only lints.
Strengths
- •Blocks suboptimal DDL directly in the database engine via extension hooks.
- •Visualizes byte-level padding waste clearly for developers unfamiliar with alignment.
- •Generates migration scripts to fix existing tables without manual reordering work.
Weaknesses
- •Requires superuser privileges to install extension on managed Postgres services.
- •Only solves storage padding, not index fragmentation or query planning issues.
Category
Target Audience
PostgreSQL DBAs, Backend engineers managing large-scale databases
Similar To
SQLFluff · pglint · EXPLAIN
Similar Projects
Infrastructure●●●Banger
Loomabase – Offline-first sync for SQLite and PostgreSQL
Column-level CRDTs prevent unrelated offline edits from overwriting each other.
Big BrainWizardry
vforno
201d ago
Infrastructure●●Solid
TidesDB – A persistent key-value store optimized for modern hardware
LSM-tree with SSI, column families, and adaptive compaction—solid database primitives, nothing novel.
WizardryShip It
alexpadula
1044mo ago
AI/ML●●Solid
AI agent autonomously used BoTorch to predict CFPB enforcement actions
AI agent autonomously selected BoTorch and tuned hyperparameters without human intervention.
WizardryBold Bet
mshipman
103mo ago
Productivity●Mid
Apparent – A lightweight extension to simplify and clean Gmail
Hide Gmail AI overviews and clutter, but Simplify Gmail does more for longer.
Solve My Problem
JaredCampbell
112mo ago
AI/ML●Mid
Consciousness Gateway – AI routing with consciousness-first alignment
Product Algebra routing plus an explicit 'dharma' pipeline (no-self regularization, entropy/mindfulness metrics, compassion and ethos scores) is a strikingly specific approach — it moves beyond cost/capability heuristics into cross-modal interaction scoring and reputation-driven incentives. There's real engineering here (1s perception loop, SQLite memory, Telegram UX, multi-provider SDK support), but the repo reads young and claim-heavy: I want reproducible benchmark artifacts, links from the code to the cited 439-model experiments, and clearer deployment/security guidance before trusting it for critical workloads.
Bold BetBig Brain
AIconscious
204mo ago