Note: The job is a remote job and is open to candidates in USA. Edison Smart® is establishing a new US R&D center and is looking for a Senior Machine Learning Engineer to lead the full-stack recommendation and growth ML direction. The role involves building models and infrastructure to enhance product discovery, campaign targeting, and content ranking for a platform with over 80 million users.
Responsibilities
- Design and build low-latency, real-time recommendation systems across web and mobile — owning the full ML lifecycle from data prep and feature engineering through training, evaluation, and production deployment
- Apply advanced personalisation techniques (two-tower retrieval, sequential models, GNNs, multi-objective modeling with PLE/MMoE, contextual bandits) across trading, community, and campaign surfaces, with an eye toward unifying signals across them
- Build the personalisation layer for the client's AI investment assistant: integrate portfolio signals, trading behaviour, on-chain data, and market intelligence into real-time token recommendations, market alerts, and eventually autonomous agent-driven investment workflows
- Develop predictive models for churn, upgrade propensity, reactivation, and LTV, applying causal inference (uplift modeling, difference-in-differences) to optimise intervention timing and subsidy allocation
- Build a full user lifecycle data and value system — behavioural signals predicting key conversion milestones, personalised intervention strategies, and rigorous treatment-effect measurement
- Architect and own the full-stack A/B experimentation infrastructure: assignment, metric pipelines, statistical frameworks (CUPED, sequential testing), and self-serve tooling
- Build and maintain real-time and batch feature pipelines, partnering with data engineering on feature store design and end-to-end observability
- Partner with Growth Product, Data Science, Community, and Asia-Pacific engineering teams to translate business goals into ML requirements; set engineering standards and mentor as the US team grows
Skills
- 5+ years in ML engineering, recommendation systems, or growth engineering at a consumer-scale internet company
- Proven track record shipping real-time recommendation or personalisation systems to millions of users; strong knowledge of collaborative filtering, two-tower models, sequential models, GNNs, multi-objective modeling, and contextual bandits/RL
- Solid foundation in causal inference and statistical learning: uplift modeling, A/B experiment design, treatment effect estimation
- Strong Python plus at least one JVM/compiled language (Java, Scala, Go, C++); experience with PyTorch/TensorFlow/JAX and big data tooling (Hive SQL, Spark, Flink)
- Hands-on with large-scale data infrastructure: Kafka, Spark/Flink, feature stores (Feast, Tecton, or equivalent), online serving (Redis, Cassandra)
- Some grounding in AI/Agent work — understands where recommendation systems and LLMs are converging, and has hands-on Agent-related engineering experience (anomaly detection, skill/tool integration) that can be fused with recommendation capability
- Fluent in English and Mandarin, able to bridge the US R&D center with Asia-Pacific engineering and product teams in Singapore, Dubai, and beyond (comfortable with occasional early-morning/evening cross-timezone meetings)
Benefits
- Bonus
- Equity
- Remote or Hybrid
Company Overview