Case study · AI Transcription & Meeting Intelligence
A privacy-first AI meeting transcription product: a web app plus the Axiom Meeting Scribe Chrome extension that turn meetings and recordings into speaker-labelled transcripts and AI summaries, without a bot joining the call.
The problem
Most AI note-takers work by sending a bot into the call, which announces that the meeting is being recorded and can make clients uncomfortable in sensitive conversations. Recordings were also scattered across Zoom, Google Drive and Dropbox, with no simple way to turn them into searchable transcripts, summaries and action items.
The approach
Technier designed and built Axiom as a web app and a Chrome extension. The web app ingests recordings by file upload or by pasting a Zoom, Google Drive or Dropbox link, with Google Drive connected over OAuth so files are pulled server-side. The Axiom Meeting Scribe extension records live Google Meet, Zoom, Microsoft Teams, Webex and Slack Huddles calls straight from the browser tab using Chrome's tab-capture API, with a side panel that shows recording status, so no bot joins the meeting. Audio is transcribed with Whisper Large-V3 with speaker diarization (PyAnnote), each transcript is analysed for key points, decisions and action items, and raw audio is deleted once the transcript is ready.
The stack
The results
Shipped and live: the Axiom web app with email and Google sign-in, and Axiom Meeting Scribe (v1.2.0) published on the Chrome Web Store. Users can record or import meetings, get speaker-labelled transcripts with AI summaries and action items, search across their meetings, export transcripts as PDF or DOCX, and send summaries to Slack.
More work
Autonomous AI & Workflow Automation
Built an agentic AI system simplifying enterprise hiring, automated workflow dispatch, and global talent scaling.
Tech stack: Next.js · Python · FastAPI · OpenAI
Read case study: MyWorker AI Workforce Platform
HealthTech & Cloud SaaS
Engineered a HIPAA-compliant cloud analytics platform delivering real-time patient metrics and automated diagnostics.
Tech stack: React · Node.js · AWS VPC · Docker
Read case study: Pulse Healthcare Analytics Portal
Tell us what you're building. We'll reply with clear next steps.
