LAURENCE FANG

LAURENCE FANG — INDEPENDENT WORKS




MOVE TO ENTER

01 / EVIDENCE SPACE

Can a model carry
a skill into a task?

A controlled pilot compares direct Modus Ponens probes with downstream tasks that require the same procedure in context.

The observed differences are shown directly, without extrapolating beyond the results.

READ THE STUDY

Modus Ponens evaluation · 3-model pilot · results shown as observed.

02 / SYSTEMS IN MOTION

From evidence
to agency.

Three working systems, each with an explicit operational boundary.

02 / AGENTS / CAREER TECH

Career Agent MVP

Evidence-aware job-search workflow automation.

03 / FIELD NOTES

Methods meet
the field.

Applied AI algorithm internship
at 燧人(珠海)医疗科技有限公司.
01

INTERNSHIP TRACK

Visual Classification Model Reproduction & Evaluation

EfficientNetV2 98.10% · StarNet 98.05%

Built and adjusted local model-training and evaluation workflows across dataset preparation, fine-tuning, training configuration, accuracy/F1 analysis and confusion matrices.

02

INTERNSHIP TRACK

Single-Video Multimodal Feature Extraction

TEXT 50 × 768 · AUDIO 925 × 25 · VISUAL 232 × 177

Built a local tool that extracts structured text, audio and visual/facial representations from one video.

03

INTERNSHIP TRACK

INN / Feature Engineering Experiments

MNIST 90.58% · OOF +1.94 PP

Reproduced and experimented with an INN classifier workflow, comparing raw pixels, HOG + PCA, patch selection and ensemble approaches.

READ FIELD NOTES

04 / HORIZON

Questions still
in motion.

Research directions and engineering notes, held to their actual evidence.
ENTER RESEARCH NOTES

OPEN CHANNEL / COLLABORATION & OPPORTUNITIES

Let's make the
signal useful.

I'm interested in AI engineering, research engineering, multimodal AI, LLM evaluation and agent-system opportunities.

05REPLIES WELCOMEADES.TOP