AskEcho

AskEcho.cv: A voice-enabled AI Digital Twin built with Groq and Ola.cv. It syncs with your live GitHub activity to pitch recruiters.

Comment

AskEcho.cv

Inspiration

The motivation for AskEcho.cv came from a single frustration: the "Resume Void." As a Software Engineer, I noticed that traditional PDFs are frozen in time. I wanted to bridge the gap between a static identity and a living engineering workflow. I was inspired by the idea of a Digital Twin—a professional doppelgänger that can "vibe" with recruiters and prove its worth through live data while I’m busy coding.

What it does

AskEcho is a voice-enabled, sub-second AI Digital Twin.

  • Conversational Identity: Users chat with the agent about my background; it responds with a human voice and a reactive Vibe Visualizer.
  • Live Pulse: A real-time GitHub activity feed allows the AI to discuss my most recent commits as they happen.
  • Instant Intelligence: It uses the fastest inference on the planet to ensure zero recruiter wait times.

How I built it

The project is built on a high-performance stack optimized for the "Vibe Coding" era:

  • Brain: Groq (Llama 3) handles the reasoning. We optimized for a total response latency where:

  • Identity: Integrated with the Ola.cv Developer API for programmatically managed decentralized identity.

  • Voice: Browser-native Web Speech API paired with a custom CSS "Breathing Orb" visualizer. The orb's scale follows a simple harmonic motion when active:

  • Data: A backend integration with GitHub (Octokit) to pull real-time commit metadata.

Challenges we ran into

The biggest hurdle was Voice Latency. Using cloud-based TTS made the "Twin" feel laggy. I pivoted to a client-side Web Speech API implementation. This required building a state-management system to sync the visualizer animations with browser audio events. Additionally, managing domain delegation via a Python script and the Ola.cv API (rather than a GUI) was a steep learning curve that proved the power of "Identity-as-Code."

Accomplishments that we're proud of

  • Identity-as-Code: Successfully automating the registration of askecho.cv via the developer API.
  • Sub-Second Vibe: Achieving a near-instant response loop that makes the AI feel like a real person, not a slow chatbot.
  • Dynamic Context: The AI actually "knows" what I pushed to GitHub five minutes ago.

What we learned

I learned that speed is a feature. A fast, responsive system provides a better "vibe" than a complex, slow one. I also gained deep experience in the Ola.cv protocol, realizing how human-readable names are the future of professional decentralized identity.

What's next for AskEcho

The goal for AskEcho is to become a full-scale Career Agent:

  • Active Outreach: Training the agent to proactively notify me when a recruiter views my profile.
  • Live Project Demos: Integrating a feature where the AI can "spin up" a sandbox demo of my projects right in the chat.
  • Ola.cv Expansion: Helping more Nigerian engineers automate their professional identities via the API.

Built With

  • github
  • groq
  • next.js
  • octokit
  • ola.cv
  • tailwind
  • webspeechapi
Share this project:

Updates