Principal Applied AI Engineer (moment)

Overview

This role focuses on helping people learn to cook in innovative ways via a personal coaching service powered entirely by AI, leveraging multimodal inputs (text, vision, audio). Rather than incremental improvements, the position calls for identifying and solving the most impactful problems that narrow the gap between today's AI and the ideal learning experience.

Key Challenges

• Video Analysis: Emulate top cooking coaches' judgment via AI, discerning heat levels, signs of failure, and appropriate corrections from video.
• Fundamental AI for Cooking Videos: Solve core challenges in interpreting and reasoning about cooking videos.
• Multimodal Modelling: Design decision-making models that use video, audio, and text together.
• Coaching Logic: Create systems that factor in conversational memory and user learning state.
• Research & Task Agents: Implement agents for both research and product features.

Responsibilities

• Start with the ideal user learning experience, identify the most critical problems, and independently define the solution path.
• Translate challenges into technical requirements and lead the design, implementation, and validation.
• Explain and measure how your solutions impact learning speed, retention, reproducibility, and cost.

Requirements (Must-Haves)

• ~5+ years software development experience.
• Professional-level Python proficiency.
• Direct experience developing and integrating with large language models (LLMs).
• Demonstrated ability to define and focus on the single highest-impact problem from many possible improvements.
• Experience with rapid hypothesis–design–implementation–validation cycles for unsolved problems.
• Practical skills in optimizing generative AI output.
• Understanding of modern software development (TDD, version control, CI/CD, containerization, etc.).
• Ability to translate nuanced user context and emotion into technical requirements.

Preferred Qualifications

• Excellent English communication in cross-functional teams.
• Experience analyzing user or conversational data to inform AI design.
• Background in computer vision or video data processing.
• Hybrid development across generative AI and traditional ML methods.
• Sensitivity to human–AI interaction and user experience.
• Knowledge of AI model trade-offs and constraints.
• Experience with vector embeddings and databases (e.g., Pinecone).
• Familiarity with Retrieval Augmented Generation (RAG) architectures for LLM-based apps.
• Experience fine-tuning/customizing large language models (LangChain, LlamaIndex, Hugging Face, AWS Bedrock, Google Vertex AI).
• Development experience in cloud (AWS, GCP, Azure).
• Understanding of MLOps / LLMOps (deployment, serving, monitoring, etc.).
• Foundation in data engineering and building data pipelines for AI.

Location

Ohashi, Meguro-ward, Tokyo
(Flex time / no core time)

Compensation & Benefits

• Annual opportunity for salary revision based on biannual evaluations
• Allowances: Visa support for overseas employees; company housing (with conditions)
• Insurance: Unemployment, industrial accident, health, pension
• Holidays: Five-day workweek (Saturday & Sunday), national holidays, year-end/New Year, paid & special leave (condolence, marriage, spousal childbirth, bereavement), refreshment leave, summer vacation (3 days)

---