Principal AI Engineer, Media Pipelines

Liquid Studios AI

27d•$150,000 - $175,000

About The Position

Liquid Studios turns existing video, audio, and text catalogs into platform-native short-form clips, multilingual dubs, and formats that AI search and assistants can find and cite. It operates as a managed service combining multi-model AI orchestration, agentic pipelines, and enterprise-grade quality control. The Principal AI Engineer is a senior individual contributor responsible for the architecture and technical direction of the core content transformation pipeline. The role partners closely with the Lead AI Engineer and cross-functional leads to establish technical direction and architectural standards, while also contributing substantial code and delivering high-priority features from prototype through production. The role addresses the challenges of orchestrating video, speech, and text models across large catalogs; controlling inference cost and latency at scale; and evaluating whether generated dubs and clips meet quality standards. It owns pipeline architecture, model and vendor strategy, processing economics, evaluation and quality controls, enterprise reliability, rights and consent safeguards, and engineering standards.

Requirements

Typically 10 or more years of software engineering experience, including technical ownership of a production system with real users, uptime commitments, and cost constraints.
Experience shipping generative AI systems to production, such as LLM or multimodal systems, agentic workflows, multi-step pipelines, or retrieval-augmented generation.
Experience addressing production concerns including evaluations, retries, failure modes, and cost.
Deep experience in Python and/or TypeScript.
Experience with at least one major cloud platform.
Strong grounding in distributed systems and data, including job orchestration, queues, storage, and databases for large, long-running, failure-prone workloads.
Experience setting technical direction for systems used by other engineers through design documents, reviews, and mentoring.
Ability to work remotely, set priorities against company goals, and keep others informed without day-to-day oversight.
Strong collaboration skills, receptiveness to feedback, and respect for technical and creative disciplines.
Execution-focused approach that connects engineering work to commercial utility and user impact.
Comfort making and revising technical decisions amid uncertainty.

Nice To Haves

Experience with media processing, including FFmpeg, codecs and containers, transcoding, loudness normalization, and caption or subtitle formats.
Experience with speech and audio machine learning, including speech recognition, text-to-speech, voice cloning, speaker diarization, or lip-sync.
Experience with video understanding, including scene detection, highlight and clip selection, or multimodal embeddings.
Experience self-hosting open-source models on GPUs, including capacity and cost planning.
Experience with retrieval and vector search over large content catalogs.
Experience in media, publishing, podcasting, or creator platforms, or selling to enterprise media clients.
Prototypes of AI tools, agents, or model experiments that can be demonstrated.

Responsibilities

Set the technical direction for media ingestion, transformation, quality control, and delivery; make and document decisions about orchestration, data models, and service boundaries.
Determine what to build, buy, or fine-tune across language models, speech recognition, text-to-speech, translation, and video models; run structured model comparisons and make the pipeline adaptable to provider changes.
Manage processing cost, latency, and throughput, including unit economics per minute of media, GPU and API spending, batching, caching, routing, and graceful degradation during provider outages or delays.
Build evaluations for dub naturalness and synchronization, translation accuracy, clip selection, and brand compliance; create automated quality gates supported by targeted human review.
Define service-level objectives, observability, incident response, and data handling for client catalogs, including unreleased or rights-restricted content.
Work with leadership on voice-cloning consent, likeness, content rights, and AI disclosure, and incorporate requirements into the pipeline.
Establish engineering standards for design documentation, code review, and testing; mentor engineers through pairing and review.
Write a meaningful share of critical code and take high-priority roadmap features from prototype to production.
Learn the pipeline end to end and ship at least one high-priority roadmap feature during the first 90 days.
Review pipeline scalability at 10x volume and document a plan to address bottlenecks during the first 90 days.
Run a structured model bake-off and change a production model choice based on its results during the first 90 days.
Reduce cost per processed minute and demonstrate the sources of savings within the first 12 months.
Ensure every output type has automated quality gates and a tracked evaluation score, and make it possible to add models, languages, or output formats in days rather than weeks.
Provide enterprise clients with clear SLOs, incident reporting, and safeguards against silent failures.