Semantics Architect - Data & AI Platform at TymeX

London, England, United Kingdom -

Full Time

Start Date

Immediate

Expiry Date

02 Mar, 26

Salary

0.0

Posted On

02 Dec, 25

Experience

5 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Yes

Skills

Semantic Architecture, Data Modeling, AI Workloads, Metadata, Data Mesh Principles, Analytics, Machine Learning, AI Agents, Data Governance, SQL, Databricks, Delta Lake, Spark, Ontology, Communication, Stakeholder Alignment

Industry

Software Development

Description

About the Role TymeX is building one of the most advanced Data & AI platforms in digital banking. A core pillar of this platform is a unified semantic layer that ensures every data product across our domains is interoperable, trustworthy, and AI-ready. As a Semantics Architect, you will design and govern this semantic backbone across domains and countries. Your work will directly power dynamic customer 360, conversational analytics, accurate LLM retrieval, and Actionable AI agents across the bank. This is a rare opportunity to define a semantic architecture from the ground up to safely unleash AI Agents and automation at Industrial Scale. What You’ll Do1. Own and Evolve the Semantic Architecture for the Bank Design, maintain, and govern AI ready semantic architecture and underlying technical framework. Working with domain experts and architects, establish naming conventions, ontologies, and data modeling rules & taxonomies. Build and maintain a canonical model that enables dynamic reconstruction of entire customer journeys across our products and channels. Define the domain event ontology and event schemas ensuring consistent semantics across all transactional, behavioral, and system events. Determine a robust metadata architecture and framework and ensure seamless integration with our foundation data platform Databricks. Define and manage the semantic runtime, ensuring semantic definitions are performant, executable, testable, queryable, and consistently applied across ETL, feature pipelines, LLM retrieval workflows, MCP Server and real-time APIs. Design the semantic graph (entities, relationships, hierarchies, business rules) that forms the backbone for reasoning, AI retrieval, event linking, and customer journey reconstruction & ultimately AI taking actions. 2. Semantic Computational Governance Framework Develop a framework to embed semantics capture through data product development in a frictionless and consistent manner. Develop intelligent AI assisted solutions to ensure semantic consistency with Data Product or Domain SME in the loop. Develop solutions to detect and remediate semantic drift. Own the product roadmap for semantic tooling & frameworks - such as semantic catalog, semantic contract validator, lineage resolver, semantic drift detector, and ontology editing/visualisation tools, canonical model builder. 3. Architect Shared Data Products Define the Semantic and Data Architecture for “Shared” Data Products such as Master Data, Reference Data, Feature Store, Regulatory Data Products, Core Product data such as User, Session, Channel. Ensure interoperable, distributed ownership aligned with Data Mesh principles wherever practical. 4. Enable AI Usecases at Industrial Scale Design semantics layer to power Classic and Modern AI (LLM) large scale workloads and usecases. Define semantic grounding strategies for LLMs, enabling hybrid retrieval (SQL + vector + ontology), context resolution, and reliable semantic augmentation for AI Agents. Develop semantics layer that powers reliable and trust worthy conversational analytics e.g. AI Genie in Databricks or custom AI Conversational Data Agents. Enable frameworks to ensure efficient and effective semantic consumption as part of Context Engineering and Machine learning. 5. Empower Product Teams Coach DPOs, Stewards, Architects, and Engineers on semantic best practices. Define templates, frameworks, and accelerators to platform teams for automation. Drive reusable semantic building blocks across product teams. 6. Partner Across Engineering, Data & Product Collaborate with Platform, Governance, and Engineering teams to turn your ideas into reality. Ensure semantics integrate with Databricks, Unity Catalog, DLT, dbt, and AI tools. Align semantic models with quality, security, lineage, and access policies. What You’ll Bring Masters degree in a relevant field. 5+ years of hands on experience in semantic or ontology or data modeling. Proven experience delivering semantics and metadata at scale for large AI workloads. Experience with hyper-scalar data platforms, semantic tools, dbt, LookML, MDX, Delta Lake, Spark, SQL, Palantir. Strong understanding of data contracts and data quality frameworks. Strong understanding of Data Mesh Principles and a “taste” for implementing the “Socio-Technical” changes required to democratize data Experience building analytics, ML, or AI-driven systems with semantics inputs. Hands on builder-leader who is not shy of building frameworks to accelerate understanding and development process and turn ideas into reality. Strong communication and stakeholder alignment skills. Passion for modern AI and reusable data ecosystems. Why You’ll Love Working Here Performance bonus up to 2 months. 13th month salary (pro-rata). 15 days annual leave + 3 sick days + birthday leave + Christmas leave. Meal & parking allowance. Full salary & benefits during probation. Premium healthcare for you and your family. Growth pathways, conferences, and continuous learning. International, values-driven culture. Overseas travel opportunities. Internal hackathons, team events, Blue Card program. 40-hour week, Monday–Friday. Relocation offered to Vietnam. Strong candidates overseas will be considered for remote location as well.

Responsibilities

The Semantics Architect will design and govern a unified semantic layer for the bank's Data & AI platform, ensuring interoperability and trustworthiness of data products. This role involves developing a semantic computational governance framework and architecting shared data products to enable AI use cases at scale.