Caio Theodoro Caio Theodoro

ReconForge

Laptop-trained 1.7B LoRA for financial reconciliation exceptions — 0.913 severity-weighted recall against a frontier model's 0.872, 1.000 recall on high-severity cases, zero API cost.

AIEvalsFine-tuning

Suture

8B VLM that diffs an underwriting binder against the issued policy: 0.959 recall across 13 discrepancy classes where zero-shot GPT-5.6 vision scores 0.373.

AIVisionFine-tuning

LossBench

Scores money-touching agents on expected operational loss instead of accuracy — severity-weighted, calibrated, replayable, on a hash-chained decision ledger.

AIEvalsBackend

Habeas

Form I-9 compliance validation fine-tune that cites the violated clause in M-274 / 8 CFR 274a.2, trained on synthetic-only data.

AIFine-tuning

Substrate

Five agent-systems research units, each a reference implementation paired with the benchmark harness that can falsify its own claim.

AIEvalsResearch

Specula

FDA food-label compliance review: label photo in, cited PASS/FLAG report out, grounded in 21 CFR 101 and openFDA enforcement data.

AIVisionFine-tuning

Plumb

AIA G702/G703 pay-application audit that catches math errors, retainage drift, and double-counted change orders, returning corrected figures per line.

AIFine-tuning

Perfectman

Social simulation where AI personas notice unevenly, reply late, lurk, and form private alliances — presence over tick loops.

AISimulationTypeScript

Prompt Sharpener

Tauri overlay on a global shortcut that rewrites a rough prompt into task, scope, constraints, and success criteria — Anthropic, OpenAI, Gemini, or local Ollama.

RustDesktopAI