AI & Data Solutions

Aufgaben

Objective of Job

We are seeking a hands-on, delivery-focused AI/ML Solutions expert to join our central AI Hub. Your mission is to own the end-to-end execution of AI/ML initiatives—guiding them from business concept to a robust, production-grade solution.

You will act as the critical bridge between business stakeholders and technical engineering teams, taking full ownership of the delivery lifecycle. You will validate ideas, engineer complex solutions, deploy them reliably, and continuously optimize them post-launch to ensure they deliver measurable business impact. Success is measured by the number of use cases taken to production, post-launch business KPI uplift, BU self-service adoption, and zero compliance or data-security incidents.

Task Description

Pre-Production (Feasibility & Validation)

• Requirements Translation: Convert ambiguous business requests into clear technical specifications (FRDs). Evaluate use cases for technical feasibility, data readiness, and ROI to ensure we only pursue viable projects.
• Solution Design & POC: Match business problems to the optimal technical approach (e.g., AI Agents, RAG knowledge retrieval, fine-tuning, or distillation). Lead rapid, low-cost POCs to de-risk projects, validate hypotheses, and estimate Total Cost of Ownership (TCO).
• Guardrail Definition: Establish clear behavioral boundaries for AI systems—defining knowledge sources, decision logic, fallback mechanisms, and iteration protocols before engineering begins.

Build & Engineering (Ontology & Development)

• Knowledge & Data Engineering: Architect the enterprise knowledge ontology and semantic layer to unify disparate business concepts and data silos. Structure prompt libraries and knowledge bases for optimal retrieval and performance.
• Hands-On Model Development: Lead the technical delivery of LLM/DL/ML solutions. Perform data debugging, algorithm customization, prompt engineering, and system optimization to ensure solutions meet latency, accuracy, and scalability requirements.
• Packaging for Production: Collaborate with IT to encapsulate knowledge, prompts, guardrails, and evaluation metrics into reusable, production-ready Agent skills and APIs.

Post-Production (Optimization & Continuous Improvement)

• Model Fine-Tuning & Optimization: Own the iterative performance enhancement of deployed models. Drive optimization efforts where the business case justifies them.
• Monitoring & Feedback Loops: Implement robust online evaluation and monitoring systems. Track behavioral compliance, business KPIs, and user feedback. Establish a "feedback flywheel" to continuously re-train and refine models based on real-world usage.
• Platform Advocacy: Provide critical feedback to internal IT and platform teams based on your hands-on delivery experience, helping to evolve internal tooling and infrastructure to remove bottlenecks.

Enablement (Scaling & Knowledge Transfer)

• BU Technical Support: Provide expert hands-on technical support to Business Units (BUs) for their self-driven use cases, assisting with custom debugging, hyperparameter tuning, and solution troubleshooting.
• Standardization & Self-Service: Standardize the toolchain and document vetted, reusable solutions. Build BU-specific "Golden Path" tutorials and playbooks to enable internal teams to build and scale their own AI solutions confidently and autonomously.

Governance & Regulatory Stewardship

• Regulatory Compliance, AI Governance, and Internal Policy Alignment Ensure every delivered solution rigorously adheres to all applicable financial regulatory requirements (e.g., PBOC, NFRA mandates), China's Data Security Law, PIPL, and strict internal AI Governance & compliance policies, in close cooperation with Legal, Compliance, Information Security and the Data Protection Officer.

Qualifikationen

Profile:

A pragmatic, execution-focused technologist who is comfortable diving deep into data pipelines and model weights and discussing technical constraints with business leaders, and thrives on turning AI/ML concepts into reliable, working enterprise solutions.

Qualifications

• Education: Bachelor’s degree or above in Computer Science, Data Science, Engineering, or a related quantitative field. An MBA or business background is a strong plus but not mandatory.
• Experience: 4–7 years of relevant experience with a proven, demonstrable track record of delivering AI/ML products or solutions from conception through to production deployment. Experience within cross-functional, matrixed organizations is essential.
• Technical Stack: Proficiency in Python and SQL; hands-on experience with big-data processing (e.g., Spark), MLOps tooling (experiment tracking, orchestration, model registry) and production deployment. Business-fluent Chinese and English.

Core Competencies

• Lifecycle Ownership: You possess deep, practical experience navigating the full AI project lifecycle (POC → Production → Optimization). You understand the pitfalls of each stage and know how to mitigate risks proactively.
• Technical Depth: You have strong hands-on experience with LLM/DL/ML delivery patterns (RAG, Agents, Fine-tuning, Distillation), as well as classical ML and optimization methods. You can debug data issues, optimize algorithms, and make pragmatic trade-offs between performance, cost, and speed.
• Bridging the Gap: You are able to distill ambiguous business language into rigorous, actionable technical requirements that engineering teams can execute without confusion.
• Execution Mindset: You are biased toward action and results. You remove blockers, drive decisions, and ensure that projects cross the finish line, delivering clear value rather than getting stuck in analysis paralysis.

Preferred Qualifications (Bonus)

• Prior experience with Agentic workflows or composite AI system design patterns.
• Direct involvement in building data pipelines for SFT, RLHF, or model distillation projects.
• Proven success in enabling multiple business lines to scale AI solutions independently.