Skip to content
View rosscado's full-sized avatar

Block or report rosscado

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
rosscado/README.md
 ____                   ____  _
/ ___|  __ _ _   _     |  _ \(_)
\___ \ / _` | | | |    | |_) | |
 ___) | (_| | |_| |    |  __/| |
|____/ \__,_|\__, | ___|_|   |_|
             |___/ |_____|
        voice + LLM that *listen*

Website LinkedIn X Follow X Follow Email

Hi, I'm Ross — I build real‑time voice + LLM products that listen

"Listening is the product."

Founder of Say, Pi · Applied AI Engineer · Long‑time shipper of reliable software

🏆 New Frontiers entrepreneur · ⭐ 4.9★ Chrome Web Store · 👥 Thousands of active users

I turn tricky human‑computer moments into smooth, useful experiences — specializing in real‑time voice AI, LLM orchestration, and production‑grade systems that respect how humans actually communicate.

Profile Views


What I’m building now — Say, Pi

A consumer service that gives AI assistants a natural voice interface where people already work. Think Whisper STT → LLM orchestration → expressive TTS, with low‑latency streaming, multilingual support, and polite end‑of‑speech detection (we wait for your last word).

  • Agentic listening / discretionary responses so assistants talk when useful, not just when possible.
  • Hands‑free interruption and smart barge‑in that avoids stepping on the user.
  • Production‑grade: FastAPI/Node services, Postgres, auth, observability, CI/CD; graceful failover across model providers.
  • Real users: 4.8★ Chrome Web Store rating with thousands of installs; strong retention and long sessions.

Goal: be the easiest way to add respectful, multilingual voice to any LLM surface.


Not just voice — I'm an applied AI generalist

I ship end‑to‑end systems that combine LLM integrations, evals/guardrails, RAG, observability, and reliability engineering. I'm comfortable choosing trade‑offs, then measuring them in production.

🛠️ Tech Stack

Languages & Frameworks:

TypeScript Python Java Kotlin Scala Bash Node.js Vue.js Angular Nuxt FastAPI MV3 WXT HTML5 CSS3

Infrastructure & Tools:

Linux Docker Kubernetes Helm PostgreSQL Vite Tailwind

AI & Voice:

OpenAI Groq Whisper ElevenLabs Fal.ai LLM Routing


Long road, strong rails — Merative & IBM

Before Say, Pi, I spent years shipping in regulated and enterprise contexts:

  • Merative (IBM Watson Health) — Staff engineer delivering high‑availability, HIPAA‑constrained data flows and life‑sciences apps; predictable releases, auditability, and operational readiness.
  • IBM Watson — Multilingual NLP prototypes → hardened production services (incl. Arabic/Spanish); DevOps leadership, CI/CD automation; cloud migrations for enterprise clients.
  • IBM Software Group — From feature ownership to refactors of acquired SaaS; performance, security, maintainability.

Bottom line: I bring enterprise-grade reliability and operational excellence to cutting-edge AI innovation.


Work style

  • Ship small, ship often. (weekly at minimum)
  • Tight feedback loops — user engagement drives everything; iterate fast based on real usage.
  • Build in public — share early, gather feedback, improve continuously.
  • TDD & evals — test-driven development + guardrails keep UX from regressing as models change.
  • Iterative development — ship MVP, measure, refine, repeat.
  • Privacy & reliability are user‑visible features.
  • Docs & observability so teams move faster.
  • Motto: Ship > cope.

🤝 For recruiters, collaborators, investors

🎯 What I'm looking for:

  • Founding/early engineering roles — Applied AI, voice, product engineering
  • Short consulting gigs — Standing up real‑time voice/LLM systems
  • Partnerships — Putting useful software in users' hands

📈 Traction:

  • ✅ Active users with high retention & strong reviews
  • New Frontiers entrepreneur
  • ✅ Active community & production demos

🚀 My edge:

  • Big picture AI understanding — strong grasp of the AI landscape, latest developments, and emerging trends
  • Staying current — continuously tracking and integrating cutting-edge AI/ML advancements
  • Real‑time latency & barge‑in UX that actually works
  • Multilingual support — voice that works globally
  • Solving the "don't interrupt me" problem — respectful AI that waits

📬 Let's talk:

Email: [email protected] DMs open: @saypi_ai · @rosscads


Bonus geek fact: I've compiled a weekly crossword for ~15 years. Pattern‑matching and clue‑writing pair nicely with prompts, turns out. 🧩

Popular repositories Loading

  1. watchtower watchtower Public

    Forked from containrrr/watchtower

    Automatically update running Docker containers

    Go 4 2

  2. github-issues-import github-issues-import Public

    Forked from IQAndreas/github-issues-import

    Python script to import issues from one repository into another, either one at a time or all of them at once.

    Python

  3. raceready raceready Public

    Created for toolchain: https://console.bluemix.net/devops/toolchains/0fc11092-0119-4ab8-a5f0-6cbf19d20e03?env_id=ibm%3Ayp%3Aeu-gb

    Python

  4. rosscado rosscado Public

    Personal GitHub profile README - Real-time voice AI engineer & founder of Say, Pi