AI Software Engineer

该职位来源于猎聘 Role Summary The Lead AI Software Engineer will implement Generative AI (GenAI) solutions—turning promising prototypes into secure, scalable, cost-effective, and compliant production services that deliver measurable business value. Software engineers are essential to bring AI out of the lab and into sustainable enterprise delivery. They provide the engineering backbone—platforms, pipelines, guardrails, integrations, and operational rigor—that enables consistent value at scale while managing risk and change. This role will work closely with data scientists (who design, train, and evaluate GenAI models and experiments) and ensure their work can be deployed responsibly with the right quality controls, observability, and enterprise readiness. Role Summary

• Hands-on technical lead (no direct reports)
• Emphasis on technical depth and delivery over people management
• Primary coding language: Python
• Multi-cloud environment: AWS, Azure, GCP, and Alicloud Responsibilities 1) Build and ship production GenAI solutions
• Own end-to-end delivery of GenAI applications and services (e.g., copilots, document assistants, knowledge search, summarization, workflow automation).
• Drive solution architecture and implementation with production-grade standards: testing, code reviews, packaging, versioning, and API design.
• Ensure solutions are measurable and operable: defined success metrics, runbooks, and operational readiness from day one. 2) Implement Retrieval-Augmented Generation (RAG) and enterprise knowledge integration
• Design and implement RAG pipelines including:
• Content ingestion, parsing, chunking strategies, metadata enrichment o Embeddings creation, indexing, retrieval, and (where appropriate) reranking o Grounded responses with traceability (e.g., sources/citations and evidence)
• Integrate GenAI services with enterprise systems and data sources through secure APIs and connectors. 3) Establish LLMOps quality practices (evaluation + regression)
• Build evaluation harnesses and release gates:
• Golden datasets, automated regression checks, human review workflows where needed
• Define and track operational and quality metrics:
• Groundedness, harmful/incorrect output rates, retrieval quality, latency, and cost per request
• Implement safe rollout patterns:
• Feature flags, canary deployments, A/B testing, and rollback strategies 4) Security, privacy, compliance, and guardrails by design
• Embed enterprise controls into GenAI services:
• authentication/authorization, secrets management, encryption, audit logging, traceability
• Implement guardrails for common GenAI risks:
• prompt injection mitigation, sensitive data leakage prevention, policy-based routing, safe failure modes
• Partner with security/privacy/legal stakeholders to ensure compliant deployment and responsible AI practices. 5) Leverage and align with the enterprise ML platform (MLOps)
• Build GenAI products that fit enterprise MLOps standards:
• standardized deployment, environment promotion, observability, governance, operational support
• Collaborate with platform and data engineering teams to reuse shared capabilities and identify platform improvements needed for GenAI at scale. 6) Data and delivery tooling
• Work with enterprise data platforms:
• Snowflake (primary), Databricks experience acceptable
• Build CI/CD pipelines and engineering workflows using:
• Bitbucket (preferred), GitHub acceptable Technology platforms
• Cloud: AWS, Azure, GCP, Alicloud
• Language: Python (primary)
• Data Warehouse: Snowflake (preferred), Databricks (acceptable)
• CI/CD: Bitbucket (preferred), GitHub (acceptable)
• Engineering expectations: secure-by-design, automated testing, observability, cost awareness, production readiness Required Qualifications
• Lead-level, hands-on software engineering experience delivering production systems with Python.
• Demonstrated experience shipping GenAI/LLM-powered applications into production (not just demos/prototypes), such as:
• RAG-based systems, enterprise assistants, document intelligence, workflow automation
• Experience building and operating services on one or more major cloud platforms; comfortable in a multi-cloud enterprise context.
• Working experience with enterprise data platforms (Snowflake preferred; Databricks acceptable).
• Experience implementing modern CI/CD practices (Bitbucket preferred; GitHub acceptable).
• Strong fundamentals in:
• API design, testing strategy, secure coding, performance considerations, logging/monitoring, and incident readiness Preferred Qualifications
• Containers and orchestration: Docker, Kubernetes.
• Infrastructure as Code (e.g., Terraform)
• Experience with vector search systems and retrieval/reranking patterns
• Responsible AI controls and evaluation practices (risk testing, safety checks, quality monitoring)
• Enterprise identity and access patterns (e.g., single sign-on, role-based access control) Impact Your contribution and influence will include:
• Making pragmatic architecture decisions and documenting them clearly
• Setting code quality and delivery standards through hands-on execution
• Mentoring via code reviews and technical guidance
• Aligning engineering trade-offs with business value, risk controls, and operational excellence