Hygo Studio is Hiring: AI Engineer — Classifiers, Media Intelligence & Voice R&D

Typical comp: $160,000–$290,000 (median $210,000)

Hygo Studio is hiring an AI Engineer to build the intelligent layer on top of the platform’s AI-generated media — classifying it, tagging it, organizing it — and to lead R&D into AI voice/audio generation and image intelligence. This page documents the role and the AIEH assessment bundle that maps to what the JD actually screens for.

This is a real, open role at Hygo Studio (April 2026 listing). Apply at hygo.com/#careers, or take the AIEH bundle and share your Skills Passport URL alongside your application.

What Hygo is looking for

From Hygo’s job description:

Studio generates and manages massive volumes of AI-created media — avatars, images, audio, and video. We need an engineer who can build the intelligent layer on top of this content: classifying it, tagging it, organizing it, and making it searchable and useful at scale. We’re also investing in R&D for AI voice and audio generation, and image understanding capabilities that will shape the next generation of the product.

Required skills (from the JD):

  • 3+ years building and deploying ML models in production, with emphasis on classification, tagging, or content understanding systems
  • Hands-on training experience — not just using APIs. You’ve curated datasets, experimented with architectures, tuned hyperparameters, and debugged training runs.
  • Strong experience with image classification and/or computer vision (CNNs, vision transformers, CLIP, or similar)
  • Demonstrable interest or experience in voice/audio AI — text-to-speech, voice cloning, audio classification, or speech synthesis (research, side projects, or production)
  • Proficiency in Python with PyTorch or TensorFlow; comfort reading and adapting research code
  • Experience with data labeling pipelines, annotation workflows, or active learning systems
  • Understanding of model serving in production: REST APIs, batching, latency, drift monitoring
  • Familiarity with embedding-based retrieval, vector search, or semantic similarity systems
  • Bonus: diffusion models, GANs, generative audio, published research, ONNX/TensorRT for optimization

Tech stack: Python, PyTorch, Cloud GPU clusters, REST/webhook APIs, TypeScript (for NestJS integration), PostgreSQL, Redis, Temporal.

How AIEH maps this role to a Skills Passport bundle

FamilyRelevanceWhy it matters here
Python Fundamentals0.90The JD demands Python + PyTorch proficiency. This family probes language depth (decorators, generators, context managers, asyncio) that separates ML engineers who can ship production code from those who only run notebooks.
AI Output Evaluation0.85The role is fundamentally about evaluating AI-generated content quality (classifier accuracy, generation quality, drift detection). AIEH’s AOE family tests this judgment directly — exactly mirroring the day-to-day work.
Cognitive Reasoning0.75Hyperparameter tuning, architecture decisions, and debugging training runs are reasoning-under-uncertainty problems. Cognitive Reasoning predicts the diagnostic ability the JD calls out.
Big Five Personality0.55High Openness predicts willingness to read research papers and adapt experimental code (the JD asks for both). Conscientiousness predicts the careful experimental discipline that produces reproducible results.

Pillar weights for this role

  • Domain: 0.40 — Python + PyTorch + CV + voice/audio production experience is the core gate.
  • Cognitive: 0.30 — bumped above default because ML R&D is a research problem; raw reasoning ability predicts who can navigate ambiguous research papers and adapt them to production constraints.
  • AI fluency: 0.25 — AOE tests judgment on AI outputs, which IS this role’s daily work. Default weight is appropriate.
  • Communication: 0.05 — minimal. The JD doesn’t emphasize cross-functional communication; this is heads-down R&D work.

The composite score on the 300-850 calibrated scale is what Hygo can use to compare candidates without forcing them through yet another platform’s testing flow.

Two ways to apply

Apply directly: visit hygo.com/#careers and follow Hygo’s application flow.

Apply with an AIEH Skills Passport: start with the free Big Five sample (the only fully-live family at the time of this writing). When Python Fundamentals and AOE flip to active, complete those too. Share your passport URL alongside your hygo.com application.

What AIEH cannot test (Hygo screens directly)

  • Production model training experience — the JD’s hardest non-negotiable: “You’ve curated datasets, experimented with architectures, tuned hyperparameters, and debugged training runs.” This is portfolio + reference evidence; bring detailed write-ups of training runs you’ve owned end-to-end.
  • Computer vision production experience — CNNs, vision transformers, CLIP in production. Show your inference pipelines and the latency / cost tradeoffs you made.
  • Voice/audio R&D background — research papers read, side projects, production work with TTS/voice cloning. The JD accepts research or hobby evidence here.
  • GPU infrastructure operation — you’ve operated cloud GPU clusters at cost-conscious scale; bring numbers on utilization and per-inference cost reduction you’ve achieved.

The JD also asks all applicants for: example links to live production work owned end-to-end, rough traffic/usage metrics, and an estimate of how much LLM/coding-agent contribution went into those examples.

Prove you're ready for this role

Take these AIEH-native assessments to add evidence to your Skills Passport:

Try the top test free