{"id":1209576,"url":"https://alion.io/job/enexia-mlops-engineer","title":"MLOps Engineer","company":{"id":2677237,"name":"ENEXIAaitonghuishe","domain":"enexia.co.jp","url":"https://alion.io/company/enexia-2","size_band":null,"is_staffing_agency":false,"employer_type":"direct","is_intermediary":false,"listed_via":null,"ats_vendor":"HERP","truth_index":null},"role":"AI/ML","role_family":"AI/ML","seniority":null,"employment_type":null,"work_mode":"remote","remote_scope":"stated_countries","remote_scope_basis":"inferred_payroll_markers","remote_working_hours":null,"hiring_geo_confidence":"inferred","locations":[],"countries":[],"hiring_countries":["KR"],"hiring_countries_total":1,"salary":{"min":8000000,"max":15000000,"currency":"JPY","period":"year","gross":null,"usd_annual":95295},"salary_estimate":null,"experience_years_min":null,"visa_sponsorship":false,"relocation_package":false,"has_equity":false,"technologies":[{"name":"Ansible","optional":false},{"name":"CI/CD","optional":false},{"name":"Cursor","optional":false},{"name":"Docker","optional":false},{"name":"GCP","optional":false},{"name":"GitHub Actions","optional":false},{"name":"Google Cloud Run","optional":false},{"name":"Grafana","optional":false},{"name":"Kubeflow","optional":false},{"name":"Looker","optional":false},{"name":"Metabase","optional":false},{"name":"Metaflow","optional":false},{"name":"Python","optional":false},{"name":"Scikit-learn","optional":false},{"name":"SQL","optional":false},{"name":"Terraform","optional":false},{"name":"Terragrunt","optional":false},{"name":"Time Series Forecasting","optional":false},{"name":"Vertex AI","optional":false},{"name":"Agile","optional":true},{"name":"AI Agents","optional":true},{"name":"Claude Code","optional":true},{"name":"Copilot","optional":true},{"name":"Feast","optional":true},{"name":"Feature Store","optional":true},{"name":"Kubernetes","optional":true},{"name":"MLFlow","optional":true},{"name":"Progressive Delivery","optional":true}],"status":"live","first_seen_at":"2026-08-06T11:35:07Z","employer_posted_date":"2026-08-06","last_verified_at":"2026-09-25T06:07:30Z","board_verified":true,"closed_at":null,"days_open":50,"trust":{"level":"ok","repost_count":null,"flags":[],"days_open":50},"description":"Python / GCP再エネの発電量予測を本番で回すMLプラットフォームをつくるMLOpsエンジニア\nDescription\nポジションについて\nMLOpsチームは、データサイエンティストが実験を実行・評価し、予測モデルを安全かつ再現可能に学習・デプロイ・運用するためのプラットフォームと開発プロセスを提供します。自動学習・予測ワークフロー、特徴量パイプライン、モデル検証、バージョン管理、監視、電力・市場データの挙動変化への迅速な対応を支えます。（シニア）MLOpsエンジニアとして、このライフサイクルの重要な領域を担当し、研究コードと信頼できる本番システムをつなぐ役割を担います。モデル研究そのものではなく、MLプラットフォームと運用を中心としたエンジニアリングポジションです。\n主な業務\n・データサイエンティストと連携した実験設計、モデルプロトタイプの再現可能・テスト可能・デプロイ可能なワークフローへの変換\n・バッチ／準リアルタイムの学習・予測用特徴量パイプラインの設計・構築・運用\n・データ要件・学習挙動・予測出力・デプロイ判定を対象としたモデルテスト・検証フレームワークの構築\n・予測品質、データドリフト、コンセプトドリフト、特徴量スキュー、レイテンシ、パイプライン健全性の評価フレームワーク・監視基盤の実装\n・自動学習・再学習・デプロイ・ロールバック・モデルバージョン管理ワークフローの設計\n・新規モデル・新規予測対象へのスケール支援、モデルアーキテクチャ・実装のトレードオフと落とし穴の特定・共有\n・MLプラットフォームの信頼性・セキュリティ・コスト効率・可観測性の改善と本番障害対応\n・Data／Platform・SREチームと連携した共通のデータ・オーケストレーション・インフラ・ガバナンス基盤の整備\n現在の技術スタック（現時点の構成）\n言語：Python、SQL／ML・データ：pandas、scikit-learn、勾配ブースティング、時系列ワークフロー／オーケストレーション：Airflow、Metaflow、Kubeflow、GCPマネージドサービス／監視：Grafana、Looker、Metabase／クラウド：GCP、Docker、Cloud Run、Vertex AI（必要に応じて）／インフラ：Terraform、Terragrunt、Ansible／CI/CD：GitHub Actions、Cloud Build／開発：Cursor（主要IDE）、クラウドエージェント\n──────────────────────────────\n[About the Role]\nThe MLOps team provides the platform and engineering practices that let data scientists run and evaluate experiments and train, deploy, and operate forecasting models safely and repeatably. As a (Senior) MLOps Engineer, you will own significant parts of this lifecycle and act as the engineering bridge between research code and reliable production systems. This is a platform-and-operations role, not ownership of model research itself.\n- Responsibilities:\n- Partner with data scientists to design experiments and turn model prototypes into reproducible, testable, deployable workflows.\n- Design, build, and operate feature-engineering pipelines for batch and, where needed, near-real-time training and prediction.\n- Build model testing and validation frameworks (data requirements, training behavior, prediction outputs, deployment gates).\n- Implement evaluation frameworks and monitoring for prediction quality, data/concept drift, feature skew, latency, and pipeline health.\n- Design automated training, retraining, deployment, rollback, and model-versioning workflows.\n- Improve reliability, security, cost efficiency, and observability of the ML platform and respond to production incidents.\n- Collaborate with Data and Platform/SRE teams on shared data, orchestration, infrastructure, and governance foundations.\n- Current Stack: Python, SQL; pandas, scikit-learn, gradient boosting, time-series workflows; Airflow, Metaflow, Kubeflow, managed GCP; Grafana, Looker, Metabase; GCP, Docker, Cloud Run, Vertex AI; Terraform, Terragrunt, Ansible; GitHub Actions, Cloud Build; Cursor (primary IDE), cloud agents.\nRequirements\nポジションについて\nMLOpsチームは、データサイエンティストが実験を実行・評価し、予測モデルを安全かつ再現可能に学習・デプロイ・運用するためのプラットフォームと開発プロセスを提供します。自動学習・予測ワークフロー、特徴量パイプライン、モデル検証、バージョン管理、監視、電力・市場データの挙動変化への迅速な対応を支えます。（シニア）MLOpsエンジニアとして、このライフサイクルの重要な領域を担当し、研究コードと信頼できる本番システムをつなぐ役割を担います。モデル研究そのものではなく、MLプラットフォームと運用を中心としたエンジニアリングポジションです。\n主な業務\n・データサイエンティストと連携した実験設計、モデルプロトタイプの再現可能・テスト可能・デプロイ可能なワークフローへの変換\n・バッチ／準リアルタイムの学習・予測用特徴量パイプラインの設計・構築・運用\n・データ要件・学習挙動・予測出力・デプロイ判定を対象としたモデルテスト・検証フレームワークの構築\n・予測品質、データドリフト、コンセプトドリフト、特徴量スキュー、レイテンシ、パイプライン健全性の評価フレームワーク・監視基盤の実装\n・自動学習・再学習・デプロイ・ロールバック・モデルバージョン管理ワークフローの設計\n・新規モデル・新規予測対象へのスケール支援、モデルアーキテクチャ・実装のトレードオフと落とし穴の特定・共有\n・MLプラットフォームの信頼性・セキュリティ・コスト効率・可観測性の改善と本番障害対応\n・Data／Platform・SREチームと連携した共通のデータ・オーケストレーション・インフラ・ガバナンス基盤の整備\n現在の技術スタック（現時点の構成）\n言語：Python、SQL／ML・データ：pandas、scikit-learn、勾配ブースティング、時系列ワークフロー／オーケストレーション：Airflow、Metaflow、Kubeflow、GCPマネージドサービス／監視：Grafana、Looker、Metabase／クラウド：GCP、Docker、Cloud Run、Vertex AI（必要に応じて）／インフラ：Terraform、Terragrunt、Ansible／CI/CD：GitHub Actions、Cloud Build／開発：Cursor（主要IDE）、クラウドエージェント\n──────────────────────────────\n[About the Role]\nThe MLOps team provides the platform and engineering practices that let data scientists run and evaluate experiments and train, deploy, and operate forecasting models safely and repeatably. As a (Senior) MLOps Engineer, you will own significant parts of this lifecycle and act as the engineering bridge between research code and reliable production systems. This is a platform-and-operations role, not ownership of model research itself.\n- Responsibilities:\n- Partner with data scientists to design experiments and turn model prototypes into reproducible, testable, deployable workflows.\n- Design, build, and operate feature-engineering pipelines for batch and, where needed, near-real-time training and prediction.\n- Build model testing and validation frameworks (data requirements, training behavior, prediction outputs, deployment gates).\n- Implement evaluation frameworks and monitoring for prediction quality, data/concept drift, feature skew, latency, and pipeline health.\n- Design automated training, retraining, deployment, rollback, and model-versioning workflows.\n- Improve reliability, security, cost efficiency, and observability of the ML platform and respond to production incidents.\n- Collaborate with Data and Platform/SRE teams on shared data, orchestration, infrastructure, and governance foundations.\n- Current Stack: Python, SQL; pandas, scikit-learn, gradient boosting, time-series workflows; Airflow, Metaflow, Kubeflow, managed GCP; Grafana, Looker, Metabase; GCP, Docker, Cloud Run, Vertex AI; Terraform, Terragrunt, Ansible; GitHub Actions, Cloud Build; Cursor (primary IDE), cloud agents.\nPreferred Experiences\n歓迎要件（NICE TO HAVE）\n・エネルギー領域のモデリング経験（気象予測、再生可能エネルギー発電量、電力需要、コモディティ価格変動、その他の時系列問題）\n・Vertex AI、MLflow、Feast、モデルレジストリ、Feature Store、実験管理基盤の経験\n・IaCツール（Terraform、Terragrunt、Ansible）の経験\n・MLアプリケーション向けCI/CDおよび段階的リリースの経験\n・データ品質フレームワーク、リネージ、ガバナンス、再現性管理の経験\n・Kubernetes、分散処理、GPUワークロード、性能・コスト最適化の経験\n・本番MLシステムにおけるオブザーバビリティと障害対応の経験\n・電力、取引、FinTech領域の経験\n・スタートアップ、アジャイル、多文化チームでの経験\n・事業・ドメインコミュニケーションに必要な日本語力\n・Cursor、Claude Code、Copilot等を用いたAI支援／エージェント型開発経験\n──────────────────────────────\n[Requirements (Nice-to-have)]\n- Energy-related modeling: weather forecasting, renewable generation, energy demand, commodity price fluctuation, or other time-series problems.\n- Vertex AI, MLflow, Feast, model registries, feature stores, or experiment-tracking platforms.\n- IaC tools: Terraform, Terragrunt, or Ansible.\n- CI/CD and progressive delivery for ML applications.\n- Data-quality frameworks, lineage, governance, and reproducibility controls.\n- Kubernetes, distributed processing, GPU workloads, or performance and cost optimization.\n- Observability and incident-response practices for production ML systems.\n- Energy, trading, or fintech domain experience.\n- Startup, Agile, or multicultural-team experience.\n- Japanese language ability for business and domain communication.\n- AI-assisted or agentic development using Cursor, Claude Code, Copilot, or similar tools.\nWe are looking for\n求める人物像・カルチャー\nENEXIAは「できあがったチームに加わる」のではなく、エンジニアリングチームそのものをつくるフェーズにあります。設立から約2年、外部ベンダー開発のシステムの内製化を進めており、チームのあり方・開発プロセス・技術基盤を一緒につくる余地が大きく残されています。\n・AIを振り切って使う一方、モデルが答えを返した「あと」を高い基準で問える方（出力を理解し、検証・レビューし、自分の名前で世に出す。判断・意思決定・責任は人間が担う）\n・肩書きではなく仕事への向き合い方としてのリーダーシップを発揮し、担当領域に閉じず課題を自ら見つけ、必要な人を巻き込んで前に進められる方\n・アーキテクチャや運用の課題に対し、議論だけで終わらせず自ら手を動かして調査し、プロトタイプやデモで提案を具体的な行動へ変えられる方\n・好奇心と適応力があり、既存の前提（自分たちの前提を含む）を疑い、曖昧な状況を具体的なアクションへ変えられる方\n・ナレッジを個人で囲い込まず、共有・議論・実践を通じてチームの価値を高められる方\n──────────────────────────────\n[Who We're Looking For / Culture]\nWe are building the engineering team, not joining one that already exists. ENEXIA was founded about two years ago and is bringing core engineering capabilities in-house, so there is real room to shape the team, its practices, and the systems it owns.\n- Uses AI as far as it will go, but holds a high standard for what happens after the model returns an answer - understanding, testing, and reviewing output, and putting their own name behind it. Judgment, decisions, and accountability stay human.\n- Treats leadership as a way of working, not a title: looks beyond their domain, identifies problems without waiting for instructions, involves the right people, and moves work forward.\n- When they see an architectural or operational issue, investigates hands-on and builds a prototype or demonstration rather than only discussing it.\n- Brings curiosity and adaptability, challenges existing assumptions (including our own), and turns ambiguity into concrete action.\n- Shares knowledge rather than protecting it individually.\nWorking Conditions\nSalary\n給与\n年収 800万円 〜 1,500万円 ※経験・スキル・成果に応じて決定\n・給与形態：年俸制（16分割）\n・賞与：年2回（年俸16分割のうち2/16を原則6月・12月に支給。会社・個人の成果に応じ最大50%増の可能性あり）\n・昇給：あり（評価に基づく）\n・各種手当：生涯設計手当 15,000円/月、社員紹介手当（試用期間終了後 100,000円／その後毎月 10,000円）、通勤費実費支給\n・支払日：毎月20日（金融機関の非営業日の場合は直前営業日）\n──────────────────────────────\n[Salary]\nAnnual salary: JPY 8,000,000 - 15,000,000 (determined by experience, skills, and performance).\n- Annual salary system (paid in 16 installments); bonus twice a year (2/16 of the annual salary in June and December in principle; up to +50% based on company and individual performance).\n- Raises based on evaluation. Allowances include a life-plan allowance (JPY 15,000/month), a referral allowance, and commuting expenses.\nLocation\n勤務地\n東京本社（東京都千代田区・大手町）／リモート中心のハイブリッド\n・出社は週1〜2回を基本とし、火曜・木曜を推奨出社日としています\n・上限なく在宅ワーク可（フルリモートは個別事情・職務内容に応じて相談可）\n・受動喫煙対策：屋内禁煙\n──────────────────────────────\n[Location]\nTokyo HQ (Otemachi, Chiyoda-ku, Tokyo) / remote-centered hybrid.\n- Generally 1-2 office days per week, with Tuesday and Thursday as recommended days; flexible arrangements are possible based on individual circumstances and the role.\nJob Type\n正社員（無期雇用） / Full-time (permanent)\nWork hours\n勤務体系\n・専門業務型裁量労働制（コアタイムなし）\n・始業／終業時刻の目安：9:30／17:45\n・完全週休2日制\n・年次有給休暇（年次有給休暇とは別に入社日に10日の有給休暇を付与）\n・特別休暇（結婚休暇、配偶者出産休暇、忌引休暇、公用休暇、変災休暇、裁判員等のための休暇）\n・所定労働時間を超える労働：有り／雇用期間の定め：無し\n・兼業・副業：可\n──────────────────────────────\n[Work Style / Hours]\n- Discretionary work system for professional roles (no core hours); reference start/end times 9:30-17:45.\n- Two full days off per week; annual paid leave (10 days granted on the join date, separate from statutory paid leave) plus special leave.\n- Secondary employment and side projects are permitted.\nProbation period\nあり（3ヶ月）※試用期間中の待遇変更なし / Yes (3 months), no change in conditions during probation\nBenefits\n福利厚生・待遇\n◾健康・ウェルビーイング：社会保険完備（健康保険・厚生年金・雇用保険・労災保険）／年1回の健康診断（会社実施）／人間ドック補助（35歳以上・定期健診の代替として受診可、補助上限80,000円/税込）\n◾コミュニケーション・チームづくり：懇親会費用補助\n◾個人成長支援：カフェテリアプラン（12万円/年）\n◾柔軟な働き方：上限なく在宅ワーク可（出社推奨日あり）／専門業務型裁量労働制／兼業可\n◾資産形成・手当：生涯設計手当／確定拠出年金（DC）\n◾休日休暇：完全週休2日制／年次有給休暇（入社日に10日付与）／特別休暇（結婚・配偶者出産・忌引・公用・変災・裁判員等）\n──────────────────────────────\n[Benefits]\n- Health & well-being: full social insurance (health, pension, employment, workers' comp); annual company health check-up; medical check-up subsidy for age 35+ (up to JPY 80,000).\n- Team building: social-gathering subsidy. Personal growth: cafeteria plan (JPY 120,000/year).\n- Flexible work: unlimited remote work (with recommended office days); discretionary work system; side work permitted.\n- Asset building & allowances: life-plan allowance; defined-contribution pension (DC).\n- Leave: two full days off per week; annual paid leave (10 days on join date); special leave.","description_format":"text","description_chars":11038,"description_truncated":false,"requirements":{"experience_years_min":null,"management_years_min":null,"team_size_min":null,"manages_managers":false,"education":null,"security_clearance":false,"languages":[{"language":"Japanese","level":"All levels","optional":true}]},"benefits":["Cafeteria","Flexible schedule"],"hiring_locations":[{"name":"South Korea","iso":"KR","kind":"country"}],"hiring_excludes":[],"relocation_offered":false,"industries":[],"lifecycle":[{"event":"open","at":"2026-09-25T03:07:23Z"}],"liveness":{"score":14,"band":"cold","label":"Long shot","p_open":1,"p_active":0.397,"p_room":0.35,"age_days":49,"expected_fill_days":24,"reasons":["conf:2","win:tail"],"computed_at":"2026-09-25T05:45:01Z"},"pay":{"stated_usd_annual":95295,"is_top_pay":false},"html_url":"https://alion.io/job/enexia-mlops-engineer","json_url":"https://alion.io/job/enexia-mlops-engineer.json","meta":{"generated_at":"2026-09-26T04:03:09Z","cache_seconds":300,"methodology":"https://alion.io/methodology","terms":"https://alion.io/terms","contact":"https://alion.io/contact","api":"https://alion.io/developers","usage":{"tier":"crawler","counted_by":"address","units_charged":1,"used_today":4387,"day_limit":5000,"remaining_today":613,"minute_limit":60,"resets_at":"2026-09-27T00:00:00Z"}}}