Develop and improve LLM-powered chatbot and AI agent solutions for seller-facing scenarios.
Explore agent design, harness engineering, and skill design to improve the reliability, effectiveness, and scalability of AI applications.
Mine and develop reusable AI skills from seller queries, conversation logs, business workflows, and existing knowledge.
Analyze chatbot interaction logs and bad cases, identify failure patterns, and improve prompts, skills, tools, workflows, and agent decision-making.
Explore self-improvement agent approaches that can automatically evaluate, diagnose, and iteratively optimize chatbot capabilities.
Design evaluation frameworks and experiments to measure chatbot quality, task completion, accuracy, and business impact.
Work closely with product managers, business teams, and engineers to understand seller needs and translate them into practical AI solutions.
Research and Develop Algorithms: Identify various fraud orders in e-commerce search and recommendation, design and develop effective risk control strategies and models to guarantee user experience and seller benefit.
Model Optimisation: Explore, innovate, and optimise risk control models to improve precision, recall, and scalability.
Data Analysis: Conduct in-depth analysis of business data, user behaviour data, and seller/item information, leveraging various anomaly detection methods for modelling.
Business and Algorithm Integration: Align risk control strategies with advertising business objectives, establish a robust anomaly monitoring system, and quickly detect and address risks.
Industry Trends Monitoring: Stay updated on the latest trends in risk control and anomaly detection technologies, exploring innovative algorithms.
Bachelor's degree or higher in Computer Science or a related field
Proficiency in at least one programming language, such as Python, C++ & Go
Skilled in at least one mainstream deep learning framework, like PyTorch, TensorFlow
Passion for LLM-related technology and solving challenging real-world problems
Deep understanding of computer fundamentals, including data structures, algorithms, and machine learning
Familiarity with mainstream algorithmic models in large language models, natural language processing, recommendation, and multimodality (including principles and implementation)
Prior research or internship experience in natural language Q&A or recommendation technology is preferred
You will join our Marketplace Intelligence & Data team to work on real-world identity verification and AI security problems, including:
Develop and improve computer vision and multimodal models, such as:
AIGC detection and media forensics
Face intelligence, liveness detection, and deepfake defense
Document understanding and fraud detection
Multimodal reasoning and AI agent
Participate in the full model lifecycle, including data analysis, model training, evaluation, and deployment
Design and implement training and inference pipelines using modern deep learning frameworks
Collaborate closely with senior engineers and product managers to deliver production-ready AI solutions
Continuously learn and explore new techniques in computer vision, multimodal AI, and trustworthy AI
Develop distributed Parameter Server (PS) systems for large-scale sparse model training and inference platforms in the search, advertising, and recommendation domains. The system should support high-throughput parameter read/write and update operations, handle hundreds of billions of features and TB-level sparse models, enable online real-time learning, and meet algorithmic needs such as feature admission and expiration.
Participate in the development of the one-stop machine learning platform, integrating the PS system into the platform to provide a user-friendly, stable, high-performance, and platform-level distributed parameter service system. Enhance the platform’s efficiency and usability, accelerating the model iteration process for algorithm teams.
Develop distributed Parameter Server (PS) systems for large-scale sparse model training and inference platforms in the search, advertising, and recommendation domains. The system should support high-throughput parameter read/write and update operations, handle hundreds of billions of features and TB-level sparse models, enable online real-time learning, and meet algorithmic needs such as feature admission and expiration.
Participate in the development of the one-stop machine learning platform, integrating the PS system into the platform to provide a user-friendly, stable, high-performance, and platform-level distributed parameter service system. Enhance the platform’s efficiency and usability, accelerating the model iteration process for algorithm teams.
In a data-driven and evaluation-driven manner, build an efficient closed loop for data iteration and establish an end-to-end data system spanning data sourcing, labeling, processing, synthesis, and evaluation. Continuously build high-quality datasets and evaluation sets to keep improving foundation model capabilities and to drive the development of AI models and applications.
Responsibilities cover one or more of the following directions:
Design and implement high-performance, scalable, and distributed data infrastructure covering the full lifecycle — data storage, ingestion, cleaning, labeling, management, and analysis — and continuously improve data engineering efficiency.
Design audio-visual multimodal training data strategies; develop efficient data processing, synthesis, and optimization operators and pipelines, and build a multimodal data asset repository to meet the data needs of large model development.
Build a "data–model–evaluation" closed loop together with Agents, using data to drive rapid iteration of large models.
Track cutting-edge techniques and methods in the large-model data domain, explore innovative approaches such as data augmentation, dat
Own the architecture, development, and productionization of core AI data product capabilities, including natural language to SQL, dashboard generation, and insight / report generation.
Build production-grade LLM application pipelines for enterprise data scenarios, including:
metadata and schema retrieval
semantic layer and metric understanding
query planning and SQL generation
SQL validation, rewriting, and execution
visualization spec generation
narrative insight generation
Design and implement reliable backend services and platform capabilities that balance accuracy, access control, explainability, observability, latency, and cost.
Deeply understand user workflows in analytics and data consumption, and translate ambiguous business questions into scalable, reusable engineering systems.
Work closely with Data Infra, Data Engineering, BI, Frontend, Design, and AI / ML engineering teams to deliver products from zero to one and scale them from one to many.
Establish and continuously improve the LLM evaluation and feedback loop, including offline benchmarks, online metrics, user feedback, prompt / model versioning, failure analysis, and quality improvement.
Drive performance, reliability, and engineering excellence across the system, and contribute to long-term architectural evolution.
Stay current with best practices in LLM applications, AI agents, semantic analytics, and enterprise AI systems, and turn them into practical production solutions.
Architecture Design: Lead the architecture design and implementation of video generation post-training, focusing on high-quality instruction data, preference alignment, and video quality enhancement.
Capability Expansion: Explore long-video modeling, storyline consistency, and precise camera control.
Alignment & Evaluation: Build video quality evaluation pipelines and multi-dimensional Reward Models; execute alignment training using RLHF, DPO, GRPO, and PPO.
Inference Acceleration: Research and implement model distillation and other techniques for Diffusion model inference acceleration.
Agentic Workflow: Design and implement core Agent architectures and Agent tooling
API integration & Deployment: Build and optimize API’s for Agent lifecycle and deployment for runtime access.
Memory & Tools: training optimization in Agent and Tool-use directions, improving metrics such as tool-calling accuracy, multi-turn instruction following, and hallucination control.
Data & Evaluation: Manage the collection, cleaning, synthesis, and annotation workflows for post-training data; participate in annotation guideline development and vendor coordination.
Design and optimize a consumer-facing e-commerce Shopping Agent, covering intent understanding, task planning, query rewriting, product search/recommendation, tool calling, memory, and multi-turn dialogue.
Develop and implement post-training pipelines such as SFT, DPO, PPO/RLHF, and Reward Modeling, covering data construction, training, evaluation, deployment, and regression.
Improve tool-use accuracy, instruction following, product relevance, factual consistency, personalization, and hallucination control through data and model alignment.
Build post-training datasets and evaluation systems combining human evaluation, automated metrics, and LLM-as-a-Judge.
Drive continuous optimization based on offline evaluation, A/B testing, user experience, conversion, latency, cost, and stability.
Collaborate with product, engineering, search/recommendation, and operations teams to deploy Agent capabilities in large-scale e-commerce scenarios.
Design and develop core agent algorithms, including multi-turn dialogue planning, tool orchestration (retrieval, ranking, LLM synthesis), and adaptive task planning.
Build and maintain agent memory architectures using knowledge graphs to support long-term consistency, personalization, and context retention across sessions.
Develop emotion-aware dialogue modeling techniques to improve agent naturalness, consistency, and user engagement.
Design and implement LLM alignment and safety strategies (e.g., SFT, DPO) to mitigate risks such as implicit persuasion or psychological manipulation in personalized generation.
Build multimodal agent capabilities that integrate vision-language reasoning with dialogue planning for tasks such as tutoring or adaptive guidance.
Benchmark and deploy LLMs/VLMs across GPU clusters and cloud environments (e.g., AWS); build reproducible evaluation pipelines to support model and architecture selection.
Track frontier research in agent algorithms, contribute to publications, and represent findings at top-tier AI/NLP venues.
Build core Agent logic, including but not limited to task planning and orchestration, tool calling, multi-turn dialogue management, memory, RAG, context engineering, and multi-agent collaboration.
Lead Continuous Pre-training and Post-training for vertical domains and business scenarios, including building high-quality datasets and data pipelines, designing RL reward models, improving instruction following and reasoning capabilities, task completion, role-playing, anthropomorphic and personalized dialogue, proactive/reactive immersive multimodal conversation experiences, and enhancing the model's IQ and EQ.
Build long-term and short-term memory architectures, addressing issues such as forgetting and attention dispersion in long contexts, and improving immersion and consistency in long-term user interactions.
Build multimodal RAG systems, including development and optimization of key modules such as recall, ranking, long-text processing, and multi-document synthesis.
Develop the Agent's tool layer, integrating external APIs and MCP such as search, code interpreters, browsers, sandboxes, and third-party services.
Design and tune prompts and context management, with tailored optimization for different product requirements.
Design scientifically rigorous quantitative evaluation systems and plans aligned with product requirements; continuously monitor product metrics and provide guidance for Agent and model optimization.
Explore innovative AI applications.
Lead continuous pre-training and post-training of large language models (Dense and MoE architectures) for vertical business domains, including domain-specific data synthesis, multi-stage data curation, and monthly iteration to support downstream business integration.
Design and build autonomous agents (e.g., Plan-and-Execute agents) that automatically retrieve, extract, and synthesize domain knowledge from large-scale multilingual/unstructured data sources, producing high-confidence corpora for continued pre-training.
Develop and iterate RL-based post-training methods (e.g., CPT, SFT, DAPO) to improve model performance on domain-specific QA, reasoning, and knowledge-modeling tasks.
Build and optimize multimodal LLM applications for content generation and understanding, including product copywriting, semantic search recall, and business potential/CTR prediction.
Design end-to-end agent workflows integrating query understanding, content generation, validation, and downstream product/search systems.
Build generative representation-learning pipelines (e.g., token-based generative pre-training on behavioral sequences) to produce reusable embeddings for downstream ranking and recommendation models.
Collaborate cross-functionally to translate business requirements into scalable model training and agent system design; continuously monitor production metrics to guide model and agent optimization
Lead the design and evolution of core algorithms for:
- AIGC detection and media forensics
-Image / video quality assessment
-Face intelligence, liveness detection, and deepfake defense
-Document understanding, OCR, and fraud detection
Own end-to-end algorithmic solutions
Drive research-to-production translation
Define technical directions in
-Multimodal and vision-language models
-Trustworthy AI and biometric security
-Generative model detection and provenance analysis
Mentor senior engineers and act as the technical bar-raiser for major algorithmic decisions
Partner with product and engineering teams to align algorithm strategy with business impact
Lead the technical architecture and end-to-end development of AI Agentic products for a core enterprise platform, partnering closely with the Product Manager to shape product direction from a technical feasibility standpoint.
Architect and build scalable agent frameworks (multi-step reasoning, tool/function calling, memory, orchestration across multiple LLMs and services) that power our product's copilot and automation features.
Own the full lifecycle of AI features — technical design, model selection/fine-tuning, evaluation, production deployment, and monitoring — ensuring reliability, latency, and cost efficiency at scale.
Mentor junior AI engineers, set best practices for prompt engineering, RAG pipelines, and agent evaluation, and help build reusable AI infrastructure for the broader team.
Partner cross-functionally with product, design, backend/infra, and security teams to ensure AI features are robust, safe, and compliant with data governance requirements.
Stay on top of the latest advances in LLMs and agentic AI, and proactively identify opportunities to apply new techniques to our product roadmap.
Design, build, and ship scalable cloud-native backend services from concept to production.
Translate stakeholder needs into robust technical solutions and AI-powered product workflows, spanning estimation, design, implementation, testing, deployment, and monitoring to improve user experience, productivity, and ways of working across the organization.
Collaborate closely with backend engineers and cross-functional teams throughout analysis, development, and launch.
Leverage AI-native engineering practices, including AI-assisted development, agentic programming, and harness engineering, to accelerate delivery while writing clean, maintainable, and well-tested code with strong ownership of quality through reviews and documentation.
Plan and support infrastructure, deployment, and operational readiness for new features and services.
Mentor and coach junior engineers through code reviews, technical guidance, knowledge sharing, and hands-on support to help improve team capability and engineering standards.
Continuously improve system performance, reliability, scalability, and developer efficiency.
Design, build, and iterate on AI Agent capabilities (e.g. agent orchestration, tool/function calling, retrieval-augmented generation, multi-step workflow automation) for a core enterprise product.
Work closely with the Product Manager to translate product requirements into technical designs, quick prototypes, and production-ready features.
Develop and integrate LLM-powered features (copilot, smart search, summarization, agentic workflows) into our core enterprise product, transforming how users search, collaborate, and get work done.
Build evaluation pipelines and instrumentation to continuously measure and improve agent accuracy, latency, and reliability in production.
Rapidly prototype new ideas, run experiments, and iterate based on user feedback and product metrics.
Partner with backend/infra teams to ensure AI features are scalable, secure, and ready for enterprise use.
Independently complete major functional modules or design new products from scratch.
Execute end-to-end product design (both UX/UI) with strong UI experience.
Develop a universal design methodology by refining thought processes and accumulating experience in design practice.
Actively engage in design thinking processes to solve complex challenges and contribute innovative solutions to our products.
Stay updated on the latest market trends in design and emerging industries to keep our products competitive and relevant.
Balance multiple priorities in a fast-paced, dynamic environment while effectively managing time and delivering high-quality outcomes.
Develop a solid understanding of the front-end and back-end web development process, including design, development, and product deployment.
The Lifeline Building, Singapore 319455, Singapore (Country)
Experience
:
month - 0 month
Salary
:
TBC
We are seeking a highly motivated Frontend Engineer (up to Senior Consultant level) for a major Government project. You are not just a developer but a software craftsman who shares a strong conviction in the Government’s mission to serve the public. You will play a pivotal role, combining deep technical expertise in modern JS frameworks with UI/UX best practices to deliver secure, accessible, and high-performance digital solutions.
