Feiyu's Technical Blog
Notes on software, systems, and things I am learning.
Agents & Automation
Local models, tool-using agents, and practical information-extraction workflows.
A 1Point3Acres Assistant with a Local LLM and Codex Review
Why an LLM suggestion, permission to submit, and a verified website result need separate checks.
[2/2] How I Made a Codex-to-ChatGPT Bridge Verifiable
A small local bridge sends one authorized request and brings back a checked answer. Its architecture stays simple; the debugging reference explains the difficult boundaries.
[1/2] (Deprecated) How I Let Codex Talk to ChatGPT Through My Signed-In Chrome
The first prototype made the round trip work. Its failures taught me why browser mechanics need a smaller, testable home.
Local LLM + Agent for Web Information Extraction
A local agent architecture for semantic web extraction, browser automation, bounded model context, and durable task state.
Game Perception — Olden Era
Three ways to turn Olden Era Arena screenshots into structured game state, organized by how the system finds the icons—not by publication date.
ApproachKnown geometry · nearly fixed camera
Calibrated Grid Registration
Use the battlefield’s regular hexagonal geometry and a saved calibration to infer cell positions, then refine their alignment from screenshot edges.
- Calibrated grid
- Image edge scoring
- Small alignment correction
ApproachKnown position
Fixed-Crop Classification & Rejection
Assume each target icon appears at a known UI location. With localization removed, perception becomes a classification-and-rejection problem.
- Known UI slot
- Fixed crop
- Classify + reject
ApproachUnknown position · learned localization
Learned Object Detection
Do not assume the icon location is known. Let a standard object detector learn where icons are and what they are directly from gameplay images.
- Full screenshot
- Learned detector
- Boxes + identities
A future direction; no experiment article is available yet.
ApproachUnknown position · code-generated candidates
Sliding-Window Icon Detection
Search the full screenshot with many candidate crops, encode each crop into visual features, and match them against the known icon catalog.
- Full screenshot
- Candidate crops
- Catalog matching
Sliding-Window Icon Detection with a Catalog-Trained Backbone
Keep code-generated search windows, train the visual backbone on 390 catalog images, and learn crop quality separately. Gameplay screenshots do not update model weights.
Sliding-Window Icon Detection with a Public Vision Backbone
Search a full screenshot with code-generated candidate crops and a public pretrained visual model. No game-specific training; useful identity matches, but unreliable crop alignment.