Join Us

Role Goal

Design and engineer enterprise-grade AI Agent platforms.

Responsibilities

• Lead overall platform architecture design, including module boundary definition, microservice decomposition, API contract design, data flow design, and technology evolution roadmap;
• Design and implement multi-tenant isolation, access control, security compliance, and audit systems to meet enterprise customers' stringent data security and privacy requirements;
• Build Kubernetes, Docker, CI/CD, monitoring, alerting, and automated operations infrastructure to ensure platform stability and scalability under high-concurrency, large-data scenarios;
• Build a unified LLM invocation gateway supporting routing, retry, fallback, caching, rate limiting, token metering, and cost tracking for cutting-edge model APIs;
• Lead selection and implementation of core components including API gateway, microservice framework, message queues, databases, and caching; proficiency in Kafka, PostgreSQL, Redis, and similar infrastructure;
• Write core code with a hands-on mindset; deliver runnable, monitorable, and scalable platform versions quickly with limited resources;
• Collaborate closely with data engineers to ensure platform architecture supports data ingestion, batch-stream integration, data warehouse modeling, vectorization pipelines, and RAG data layers.

Requirements

• 1-3 years of backend or platform development experience; candidates from tier-2 internet companies, mid-size firms, or startups preferred; big-tech experience also welcome;
• Proficient in at least one of Java, Go, or Python; familiar with Spring Boot, Spring Cloud, or equivalent frameworks;
• Familiar with Docker, Kubernetes, CI/CD, and monitoring/alerting systems with hands-on deployment and operations experience;
• Familiar with Kafka, PostgreSQL, Redis, and API gateways; understands microservice architecture design principles and common pitfalls;
• Experience in microservice decomposition, multi-tenant isolation, access control, and security compliance; enterprise or government project experience preferred;
• Proven experience building platforms from 0 to 1; able to independently design technical solutions and drive implementation;
• Understanding of LLM API invocation, RAG fundamentals, and vector database concepts; ability to integrate AI capabilities into platform engineering (no model training required, but must understand AI engineering fundamentals);
• Excellent communication skills, strong sense of ownership, willing to write code at the front line; startup mindset preferred over big-tech titles;

Bonus Qualifications

• Experience with enterprise, smart city, digital government, or data middle platform projects;
• Familiarity with lakehouse technologies (Iceberg, Hudi, Delta Lake);
• Familiarity with metadata management tools (Apache Atlas, DataHub, OpenMetadata);
• Open-source contributions, technical blogs, or patents are a plus.

Role Goal

Build bridges across heterogeneous data sources for the enterprise AI Agent platform, transforming raw data into knowledge, features, and context ready for LLMs

Responsibilities

• Design and develop connectors for diverse data sources including relational databases, REST APIs, file systems, message queues, IoT devices, logs, and unstructured documents; support both batch ingestion and CDC-based real-time capture;
• Build batch-stream unified data pipelines using Kafka, Flink, Spark, DataX, etc.; implement real-time and offline ETL/ELT workflows ensuring stable data flow across ODS, DWD, DWS, and ADS layers;
• Complete data warehouse layered modeling, dimensional modeling, and metrics system construction; unify business definitions to support data services and AI applications;
• Build AI-oriented vectorization pipelines covering document parsing, text chunking, cleaning, embedding generation, vector ingestion, index updates, and version management; integrate with vector databases such as Milvus, Chroma, Weaviate, Pinecone, and pgvector;
• Construct the data layer for RAG and Agent systems, processing PDF, Word, HTML, scanned document OCR, and speech transcripts; support hybrid retrieval, re-ranking, metadata filtering, and prompt template management;
• Manage metadata and data lineage, collecting technical, business, and operational metadata; maintain version lineage at table-level, field-level, and for AI datasets;
• Define data quality rules, configure task alerting, SLA monitoring, data reconciliation, and anomaly detection to ensure AI applications rely on accurate, timely, and traceable data;
• Implement access control, data masking, audit logging, multi-tenant isolation, and sensitive information filtering to ensure sensitive data stays within boundaries and complies with security requirements for financial and government clients;
• Perform SQL tuning, job optimization, storage compression, resource scheduling, and cost control; for LLM invocations, optimize token consumption, caching strategies, and batch processing to run AI capabilities at stable, controlled costs;
• Collaborate closely with backend engineers to produce technical documentation, data standards, API specifications, and operations manuals.

Requirements

• Expert in SQL; proficient in at least one of Python, Java, or Scala;
• Familiar with Spark, Flink, Kafka, and DataX; understands batch-stream unified design principles;
• Experienced with PostgreSQL, ClickHouse, StarRocks, and similar databases with hands-on tuning expertise;
• Proven experience in data warehouse layering, dimensional modeling, and metrics system construction; able to independently complete domain modeling;
• Understanding of metadata, lineage, data quality, and access masking; hands-on data governance experience;
• Familiar with embeddings, vector databases, RAG, LangChain, or LlamaIndex; understands LLM API invocation, prompt engineering, and token cost control;
• Experience in unstructured data processing, knowledge base construction, or semantic retrieval preferred;
• Strong sense of ownership, good documentation habits, able to collaborate across teams, and willing to learn and grow quickly in a startup environment;

Bonus Qualifications

• Experience with enterprise, government, smart city, or data middle platform projects;
• Familiarity with lakehouse technologies (Iceberg, Hudi, Delta Lake);
• Familiarity with metadata management tools (Apache Atlas, DataHub, OpenMetadata);
• Proven experience building data platforms from 0 to 1;
• Hands-on RAG, knowledge base, Agent, or LLM application deployment experience;
• Open-source contributions, technical blogs, or patents are a plus.

Role Goal

Contribute to the design and engineering of enterprise-grade AI Agent platforms.

Responsibilities

• Contribute to backend service development, writing REST APIs, business logic, and basic data interfaces;
• Develop and maintain core platform modules, including user management, access control, logging, and configuration management;
• Learn and contribute to microservice implementation, gaining an understanding of service decomposition, API contracts, and basic service governance;
• Gain exposure to Docker, Kubernetes, CI/CD, monitoring, and alerting through project work; assist with basic deployment and operations tasks;
• Contribute to basic development of an LLM invocation gateway, learning about routing, retries, caching, and token metering across model APIs;
• Collaborate with data engineers on data ingestion, data services, and backend support for AI applications; write technical documentation and API specifications;
• Help deliver real departmental projects, developing business understanding in scenarios such as intelligent sales, operations, and risk management.

Requirements

• Bachelor's degree or above; Computer Science, Software Engineering, Information Management, or related majors preferred;
• Proficiency in at least one of Java, Go, or Python, with an understanding of basic data structures and algorithms;
• Knowledge of at least one backend framework such as Spring Boot, Flask, or FastAPI;
• Basic knowledge of SQL and relational databases such as PostgreSQL or MySQL;
• Understanding of basic Linux commands, Git version control, and HTTP/REST fundamentals;
• Basic understanding of, or strong interest in, LLMs, RAG, and vector databases;
• Strong learning ability; able to research and solve problems independently, proactively help where needed, communicate clearly, take responsibility, and adapt to a startup team's pace.

Bonus Qualifications

• Internship experience in backend development;
• Experience using Docker, Kubernetes, Redis, or Kafka;
• Open-source contributions, a technical blog, or GitHub projects;
• Participation in programming competitions, hackathons, or campus technology clubs;
• Knowledge of LLM API integration or frameworks such as LangChain.

Role Goal

Help connect heterogeneous data sources for the enterprise AI Agent platform and transform data into knowledge, features, and context ready for LLMs.

Responsibilities

• Contribute to ingestion from multiple data sources, including collection and cleaning of databases, APIs, files, logs, and unstructured documents;
• Develop ETL/ELT workflows and perform basic data processing using SQL, Python, or DataX;
• Assist with layered data warehouse modeling and learn the responsibilities of the ODS, DWD, DWS, and ADS layers;
• Help conduct data quality checks, configure task alerts, and reconcile data to ensure accuracy and timeliness;
• Gain exposure to vectorization pipelines through project work, contributing to document parsing, text chunking, embedding generation, and vector ingestion;
• Help build RAG knowledge bases, processing unstructured content such as PDF, Word, HTML, and OCR of scanned documents;
• Help collect and maintain metadata and data lineage, learning how to make data assets discoverable and traceable;
• Help deliver real departmental projects, developing data understanding in scenarios such as intelligent sales, operations, and risk management.

Requirements

• Bachelor's degree or above; Computer Science, Data Science, Statistics, Mathematics, Information Management, or related majors preferred;
• Proficiency in SQL, including basic queries, aggregations, and joins;
• Proficiency in Python and knowledge of Pandas, NumPy, or similar data processing libraries;
• Basic knowledge of relational databases such as PostgreSQL or MySQL;
• Knowledge of basic Linux commands and Git version control;
• Basic understanding of, or strong interest in, the big data ecosystem, including Spark, Flink, and Kafka;
• Basic understanding of, or strong interest in, LLMs, RAG, vector databases, and embeddings;
• Strong learning ability, a sense of ownership, good documentation habits, clear communication, and the ability to adapt to a startup team's pace.

Bonus Qualifications

• Internship experience in data analysis or data development;
• Experience using Spark, Flink, Kafka, or DataX;
• Knowledge of ClickHouse, StarRocks, Milvus, or pgvector;
• Experience using LangChain, LlamaIndex, or Dify;
• Kaggle or data competition experience, open-source contributions, or a technical blog;
• Exposure to OCR, NLP, knowledge bases, or semantic retrieval.

Role Goal

Drive engineering delivery of AI Agent products and build high-performance consumer applications.

Responsibilities

• Develop vertical AI Agent applications based on LLM stacks (e.g. LangChain, Qwen);
• Integrate LLMs/open-source models into product workflows and deliver production-ready modules;
• Design complex Agent workflows and optimize prompt engineering and fine-tuning processes;
• Own end-to-end AI application deployment and track LLM advances (Function Call, multimodal interaction);
• Build Agent evaluation and observability systems (tracing, tool-call chain analysis) to drive quality iteration;
• Design and iterate Agent memory and personalization for cross-session understanding and long-term companionship.

Requirements

• Recent bachelor's / master's / PhD graduates, or within 2 years of graduation;
• Bachelor's degree or above; Computer Science, Software Engineering, Machine Learning preferred;
• Strong proficiency in Python;
• Familiarity with the full LLM application development lifecycle; hands-on LangChain experience preferred;
• Able to independently debug Agent anomalies (tool-call loops, context overflow, hallucinations, etc.);
• Engineering-level understanding of prompt engineering (few-shot, chain-of-thought, structured output).

Role Goal

Design and implement the core engineering architecture and full-stack stack that powers AI Agent products, ensuring high performance, availability, and scalability.

Responsibilities

• Own overall technical architecture for AI financial agent products; lead full-stack development and integration of backend, frontend, and middleware;
• Build and maintain high-performance, low-latency real-time interaction systems for 24/7 stability;
• Develop and optimize data pipelines and knowledge systems to support industry analysis and report generation;
• Engineer and service-ize AI capabilities (model API orchestration, agent memory management) with clean, stable interfaces;
• Handle cloud deployment, monitoring, and performance optimization for large-scale user traffic;
• Build end-to-end monitoring, alerting, and debugging toolchains for production observability;
• Design async task scheduling and concurrency control for multi-user stability.

Requirements

• Recent bachelor's / master's / PhD graduates, or within 2 years of graduation; CS / Software Engineering background;
• Solid CS fundamentals; strong Python skills; familiarity with modern frontend frameworks such as React Native;
• Familiar with cloud platforms (Tencent / Azure / Alibaba) and containers (Docker / Kubernetes); microservice experience preferred;
• Strong software engineering practices; PostgreSQL design and tuning (indexes, connection pools, query optimization); API design; vector DB (pgvector) experience preferred;
• Basic understanding of AI / LLM applications; hands-on integration experience preferred; fintech interest preferred;
• Strong systems thinking and collaboration; able to own complex modules end-to-end;
• Able to independently triage production issues (pool exhaustion, cron races, WebSocket disconnects, etc.).

Internship Duration

≥3 months, ≥3 days/week

Responsibilities

• Assist in developing vertical AI Agent applications with LLM stacks (e.g. LangChain);
• Help integrate LLMs into product workflows and design Agent workflows / prompt engineering;
• Participate in AI application deployment and testing; track LLM advances (Function Call, multimodal);
• Help build Agent evaluation and observability (tracing, tool-call analysis);
• Contribute to Agent memory and personalization systems for long-term companionship.

Requirements

• Currently enrolled in bachelor's or master's programs; CS / Software Engineering / ML preferred;
• Proficiency in Python with practical project experience;
• Strong interest in LLM application development; LangChain experience preferred;
• Awareness of common Agent failure modes (tool-call loops, hallucinations) and debugging approaches;
• Basic understanding of prompt engineering (few-shot, chain-of-thought, structured output).

Internship Duration

≥3 months, ≥3 days/week

Responsibilities

• Support day-to-day product and operations work, and help write operation reports;
• Assist with user research and requirement analysis for vertical AI Agent products;
• Design growth strategies and operate communities / content platforms (e.g. Xiaohongshu);
• Analyze product usage data, produce reports, and propose optimizations.

Requirements

• Any major; Marketing / Psychology / Computer Science preferred;
• Passion for AI products; strong communication and writing skills;
• Data-sensitive, communicative, and self-driven; proficient in Excel / PPT;
• Experience in social media operations or user research preferred.

Internship Duration

≥3 months, ≥3 days/week

Responsibilities

• Contribute to product positioning and value proposition refinement;
• Help design early growth strategies and operate domestic / overseas communities and content platforms;
• Write product introductions, feature docs, and external communication materials;
• Organize online / offline branding events and explore partnerships;
• Track conversion metrics and propose optimizations.

Requirements

• Currently enrolled in bachelor's or master's programs; Advertising / Communications / Psychology preferred;
• Interest in AI products and fintech; familiarity with content platforms preferred;
• Strong bilingual (Chinese / English) content skills;
• Experience in social media, growth, or marketing preferred.