— CASE 02 / MULTI-MODEL LLM · VIVANSH INFOTECH
PDF → FORM
IN 20s.
Three LLMs in a single user flow — each chosen empirically for what it’s best at, glued together by a Pydantic-validated intermediate schema.
1h → 20sPer form · PDF → digital
200+Forms · month one
3LLMs · empirical pick
~7KTokens per form
— THE DESIGN MOVE
BUILD AN
INTERMEDIATE
JSON THAT
LLMs CAN WRITE
RELIABLY.
The platform’s native form schema was too deeply nested for any single LLM to emit in one pass — wrapper objects, type-specific aliases, radio-driven conditional containers nested arbitrarily.
The fix was a simplified intermediate JSON schema, validated by Pydantic, plus a deterministic conversion layer to the native form. That decoupled LLM understanding from structural correctness.
- PDF converter. GPT-4o reads PDF → intermediate JSON → Pydantic → native. ~20s, ~7K tokens. Bulk: 100 PDFs in 5-10 min.
- Conversational builder. GPT-4o suggests title/type; Qwen3-Coder writes blueprint; user iterates; GPT-4o-mini converts.
- Model rationale. Each picked from head-to-head testing — not vibes.
PYTHONFASTAPIGPT-4oGPT-4o-MINIQWEN3-CODER-480BTOGETHER AIPYDANTIC
— WANT THE FULL STORY?