An AI voice front desk for dental clinics
Took an AI receptionist for dental clinics from AI-scaffolded code to production, with three voice providers and per-practice tenant isolation.
Case studies first. Each covers the problem, the system, the decisions, what broke and what came of it.
Three case studies, one per kind of work: Ownership, Systems, Applied AI.
Took an AI receptionist for dental clinics from AI-scaffolded code to production, with three voice providers and per-practice tenant isolation.
Built a client backend alone: it ingests Telegram, Discord and X, summarises with AI, searches semantically and raises alerts.
Built an agent platform over an ERP: it began as expense-claim automation and became multi-company procurement and inventory intelligence.
Four more case studies.
Designed and deployed a production AWS backend for a healthcare SaaS, checked service by service against AWS's HIPAA-eligible list.
Built project skills, defined subagents and guardrails so AI agents could diagnose production safely, with a human approving every fix.
Built the API layer, PHI-access auditing features and most of the frontend for a console that operates an HL7 integration engine.
Wrote the requirements, architecture and detailed design for a multi-tenant lead SaaS, and built a demo for client review.
Code you can read.
Built an open-source, offline-first tracker that moves applications through a fixed pipeline and tailors résumés with a deterministic-first LangGraph flow.
Static Astro site with per-route metadata, structured data and CI/CD to GitHub Pages.
One line each. Work under NDA is described by function.
Moved 28 prompts from 11 workflows to a versioned backend; edits need no deploy.
Extended an HRMS (138 employee records, 2 biometric devices), then built its ERP successor.
Built the APIs, schemas and delivery pipeline around an ML cost-estimation service.
Implemented 10+ of 28 specified AI features, checked against the live OpenAI API.
vLLM behind an OpenAI-compatible proxy on one 12 GB GPU, adopted by the team.
Astro site for a Lahore café; desktop PageSpeed 100 for performance, accessibility and SEO.
Led a team of four on QGroundControl-based mission planning, operator UI and telemetry.
Peer-reviewed, Springer, 2025.
My part: Data collection, the signal-processing and feature pipeline, and model training and evaluation.
DOI 10.1007/s00034-025-03048-2 · arXiv preprint · How the study was built