Applied Research Engineer / Research Intern

Research Engineer Intern – Voice AI and Conversational Architectures

**Please note this is a remote role (Preferably Dallas or India) ideally for internship position, but open to full time consulting for the rught person.

About us

We are VoiceMonk (voicy.ai), a foundational research and intellectual property firm building the core architectures for real-time voice AI. With 14 granted patents in the United States, 2 granted patents in India, and 20 pending applications, our conversational AI research spans over a decade. We invented several of the core dialogue, streaming, and latency-reduction technologies driving today's mass adoption of multimodal voice agents.

The role

You will sit at the critical boundary between state-of-the-art voice AI research and patent engineering. You will read the latest literature on full-duplex Spoken Language Models (FD-SLMs), speech-to-speech (S2S) architectures, and real-time streaming, reproduce what matters, and build functional reference implementations of our proprietary architectures.

This is not a standard product-engineering role. You will act as the technical engine for our licensing and IP strategy—translating cutting-edge latency reduction and turn-taking models into rigorous technical specifications, evidentiary prototypes, and infringement analyses.

What you'll work on

• Reference Architectures & Evidentiary Prototypes: Design and deploy functional testbeds of our patented claims (e.g., real-time dialogue management, interruption/barge-in handling, and context retrieval). These prototypes serve as direct evidentiary validation for our licensing campaigns and defend against Alice invalidation challenges.
• Full-Duplex & Speech-to-Speech (S2S) Systems: Implement ultra-low latency voice-to-voice pipelines, streaming ASR-LLM-TTS cascades, and Voice Activity Detection (VAD) optimizations to benchmark against current industry standards.
• Technical Infringement Analysis: Analyze and reverse-engineer state-of-the-art enterprise voice agents and LLM telephony stacks to map commercial implementations directly to our existing patent claims.
• Continuation Patent Engineering: Translate architectural breakthroughs into rigorous technical whitepapers. You will work directly with our patent counsel to provide the engineering specifications necessary to file future-proof continuation patents mapping to modern LLM architectures.
• Real-Time Inference and Benchmarking: Measure Time to First Audio (TTFA), end-to-end latency across edge-cloud boundaries, and orchestration timing to prove the technical superiority of our proprietary conversational loops.

Minimum qualifications

You should meet one of the following:

• MS in Computer Science, Machine Learning, Audio Engineering, or a related field, plus 2+ years of research experience (industry research, a research lab, or published work).
• PhD in a related field, plus 1+ year of research experience.
• Currently 3+ years into a PhD program (for the internship track), with a track record of independent research.

You should also have:

• Strong Python: You should write clean, reproducible, and tested code to build robust reference architectures, not just disposable research scripts.
• Audio/ML Ecosystem Experience: Hands-on experience with modern ML frameworks (PyTorch, JAX) and the current voice AI ecosystem (streaming STT/TTS, WebRTC, LLM orchestration).
• Demonstrated Research Output: Publications, preprints, a thesis, or a reproducible body of work in speech processing, conversational AI, or low-latency ML systems.
• Self-Directed Execution: The ability to go from an academic paper (or a legal patent claim) to a working prototype independently. We will give you a technical or legal objective, not a detailed spec.

Nice to have

• Experience with full-duplex Spoken Language Models, audio codecs, or native audio-in/audio-out architectures.
• Work on inference optimization specific to voice: KV-cache management, speculative decoding, or streaming chunk orchestration.
• Familiarity with enterprise telephony infrastructure (SIP trunks, WebRTC streaming).
• Experience interfacing with legal, patent, or compliance teams on technical matters.
• Prior startup experience, or evidence that you thrive in an unstructured, highly autonomous environment.

What we offer

• Direct collaboration with the founder to shape the technical and architectural direction of a massive IP portfolio.
• A unique hybrid role combining bleeding-edge ML research with high-stakes technology licensing and patent strategy.
• Highly competitive cash compensation, supplemented by significant performance bonuses directly tied to the successful technical validation and licensing of our IP.
• Support and resources for publishing technical whitepapers and open research.
• Interns get a defined architectural project, direct mentorship, and a serious shot at a full-time offer.