Chaoran Liu

I build practical AI applications and share the architectures, prototypes, and lessons behind them.

Live demo of predictive vital measurement from browser using TensorFlow.js

less than 1 minute read

TensorFlow.js experiment

Vital estimates, right in your browser

Use your camera to explore an experimental model that estimates age, BMI, pulse, and respiration—without leaving this page.

On-deviceCamera data stays in your browser.
15 secondsHold still for one reading.
Four signalsExplore age, BMI, pulse, and respiration.
ExperimentalUse results for exploration only.

Try the live experiment

Camera permission is required. Results are estimates, not medical advice.

On-device preview

A camera becomes a sensorNo image leaves this browser.

Models are ready to load when you are.0%

Share on

Twitter Facebook LinkedIn

You may also enjoy

Beyond Summarization: Extracting Critical Fields that Matter for Insurance Decisions

11 minute read

Insurance teams often ask LLMs to summarize documents, then struggle to define what a good summary looks like. A better approach is to start from the decision, extract the critical fields, preserve their evidence, and use narrative generation only where it adds value.

From Demo to Production: Four Hard-Earned Lessons from Shipping GenAI for Insurance Claims

8 minute read

Four hard-earned lessons from turning a promising GenAI claims demo into a production-ready solution—from defining meaningful measurements and managing technical debt to building effective engineering and domain-expert feedback loops.

OCR Evaluation, Rebuilt for Insurance Claims in GenAI Era

3 minute read

Processing scanned claim PDFs used to be “good enough” if the text was mostly readable. In the GenAI era, that’s no longer true: if OCR misses a checkbox, scrambles a table, or loses layout context, the document becomes unreliable for extraction, retrieval, and automated decisions. Insurance makes this worse—many documents are old, low-quality scans with handwriting and dense forms—so we need evaluation that reflects real usability, not just character-perfect transcription.

DSPy, Meet Enterprise: Plugging in an Internal LLM Endpoint

2 minute read

DSPy is powerful, but many enterprise environments can’t call public LLM endpoints directly. This guide shows a minimal, practical way to plug an internal (non-standard) LLM service into DSPy by writing a custom dspy.BaseLM adapter.