Reception System
An AI phone handler prototype for small businesses. Answers calls, extracts intent, books appointments, using a fine-tuned LLM that can run locally with no per-call cloud fees. View the code
The Problem
Small clinics and SMEs miss calls constantly: receptionists are expensive, call-centre services are impersonal, and most AI phone tools run on a cloud subscription that doesn't suit businesses handling sensitive data. A dental clinic can't route patient calls through a third-party cloud LLM without GDPR implications.
The ask: answer calls, extract the reason, and book the appointment, with a model that can run on hardware the business already owns, no data leaving the building at inference time.
Architecture
Key Challenges
- Vanilla LLMs can't do this reliably. QLoRA fine-tuning on 600 synthetic call samples was needed before the model could produce reliable structured actions. Prompting alone was not sufficient.
- Quantisation changes behaviour. Compressing the fine-tuned models to GGUF (Q3_K_M) enables CPU-only inference but changes the quality of structured actions. This trade-off needs further evaluation before a pilot.
- Hallucination-safe slot booking. LLMs will confidently invent appointment times. Every action is validated against a Pydantic discriminated union (6 typed cases) and the calendar store before anything is confirmed: the LLM proposes, the schema and calendar confirm.
- Name capture is messier than it looks. Callers spell names out ("J-A-C-K R-E-A-C-H-E-R"), correct themselves mid-call, or get missed by regex entirely. Layered regex, spaCy PERSON fallback, and an explicit confirmation turn to catch the rest.
Outcome
A working prototype, not a live deployment: no paying client yet. Validated with 111 automated pipeline checks (booking, cancellations, edge cases, profanity handling) and benchmarked on both GPU (fine-tuned, 480 synthetic samples) and CPU-only (quantised, no GPU at inference) conditions. A separate app is currently collecting real caller audio to test against genuine speech before any commercial pilot.
Built under the UWE Enterprise Scholarship (£1,000 award), part of an MSc group project. Plan is to productise the online version as a SaaS for SMEs; the fully offline variant likely won't fit that subscription model given its zero-cloud design. Full code, eval results and test suite on GitHub