AI Platform Engineer Intern

About Hello Gubby

Hello Gubby is not an AI receptionist and not an AI answering service. It is a voice AI platform for customer operations, built on proprietary patented technology, that runs your entire front desk. Think of it as an AI workforce for lead capture, qualification, booking, and follow-up. It works across phone, website voice, chat, and video — so no lead slips through regardless of how they reach you.

Most tools offer one thing. An AI receptionist. An AI answering service. A chatbot. A video widget. Hello Gubby brings it all under one platform, powered by a single AI that knows your business inside out. The platform engages website visitors with your face through video, handles inbound calls, and re-engages customers who forgot about you. Sounds like you. Remembers like your ex, but actually useful.

The Role

We are looking for a full-stack engineering intern who lives in Python, is fluent in modern web development, and is obsessed with voice AI, LLMs, and automation tooling. You will ship production code on a platform that is already handling real customer calls every day. This is not a sandbox project. The agent you touch tomorrow is answering a real business’s phone by the end of the week.

You will work directly with the founder across the voice agent, the dashboard, the widget, and the automation layer that ties everything together. If you have ever stayed up building something with real-time voice or generative AI just to see if it would work, this role is built for you.

What You’ll Actually Do

•     Ship features on our production Python voice agent — real-time streaming STT, LLM orchestration, low-latency TTS, voice activity detection, and retrieval-augmented generation over a vector store

•     Build and extend our Next.js dashboard and the embeddable widget suite (chat, voice, video)

•     Design and deploy automation workflows: lead routing, CRM sync, follow-up sequences, analytics pipelines

•     Build and integrate against REST APIs across telephony, CRM, payments, and backend services

•     Prototype generative AI features — agent behaviors, prompt architectures, retrieval strategies, evaluation pipelines

•     Debug real production issues across the full stack: voice latency, RAG accuracy, webhook failures, frontend state bugs

•     Contribute to our multi-tenant architecture: per-customer agent configurations, industryspecific prompt modules, provider routing

•     Write code that other engineers will read, review, and build on — clean, documented, and shipped in small pull requests

What We’re Looking For

•     Currently enrolled in an engineering program (Computer Engineering, Software Engineering, Computer Science, or related)

•      Strong Python fundamentals — you are comfortable reading, writing, and debugging non-trivial Python in a production-style codebase

•     Hands-on experience with at least two of the following, and genuine curiosity about the rest:

◦          Real-time voice AI stacks (streaming STT, TTS, VAD, telephony)

◦    Generative AI and LLM tooling (prompt engineering, agent frameworks, vector databases, RAG)

◦      No-code and workflow automation platforms

◦      Next.js, React, TypeScript

◦       REST API design and integration

•     Comfort with Git, code review, and working in a shared codebase

•     Strong written communication — you can explain what you built in a PR description as well as in the code itself

•     You read documentation before you ask questions, and you verify claims against official docs before acting on them

Bonus Points

•     Exposure to Postgres, vector databases, or managed backend platforms

•     Experience with prompt engineering, agent frameworks, or multi-agent orchestration

•     Having shipped a voice AI, chatbot, or AI-powered side project that real users have touched

•     Familiarity with Tailwind, shadcn/ui, or similar component systems

•     Experience debugging real-time systems (WebRTC, WebSocket, streaming audio)

•     Contributions to open-source AI, voice, or automation projects

Why This Role Is Unusual

You will ship on the same codebase the founder is building. The feedback loop between your code and a paying customer is measured in days, not quarters. You will see every layer of a modern AI platform — from the Python agent handling a live phone call, to the Next.js dashboard rendering the transcript, to the automation workflow pushing the lead into a CRM — and you will own meaningful pieces of it end-to-end. Most interns see one slice of a tech stack. You will see all of it.