Workflow · healthcare · workflow

Incrementally improving LLM nutritional estimation in Taralli using DSPy prompt optimization

LLM-based nutritional estimation is difficult to improve without rigorous measurement; relying on subjective 'vibes' makes it impossible to know whether a change helps or hurts the system.

How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · User submits food description
The iOS app calls an endpoint that takes a food_description and returns populated messages for on-device inference.
Tools used
DSPyMIPROv2GEPAOpenRouterNutriBenchHugging FaceGemini 2.5 FlashGemini 3 FlashDeepSeech v3.2
Outcome

Switching Taralli to Gemini 3 Flash with a few-shot approach yielded ~15% accuracy improvement over the prior production model, and the system now supports fully offline on-device inference.

Results
Volume66.82%
Source

https://duarteocarmo.com/blog/from-nutribench-to-taralli-how-far-can-you-take-a-prompt

How we source this →

Grounding & classification
Source type: technical build writeup
23 fields verified against source quotes, 1 dropped as unverifiable.
predictive analyticsragbuilder submittedfailure mode describedmetric backedproduction runtime claimedtools describedworkflow describedhealthcareaccuracy improvementtechnical build writeup