LLM / GenAI Engineer — New York, NY
Зарплата не указана
ОткликнутьсяAbout The Role
The role sits at the intersection of applied ML and platform engineering: building LLM-powered systems that hold up under real production load - latency budgets, cost constraints, eval-driven iteration, and users who do not wait for the model to be right.
This team ships GenAI features that customers actually use: retrieval pipelines grounded in enterprise data, agentic workflows with tool calling, and fine-tuned models tuned for domain-specific accuracy. The work is hands-on and full-stack - prompt code to GPU infrastructure.
Key Responsibilities
• Design and build production RAG pipelines using LangChain, LlamaIndex, or custom architectures, including chunking strategies, hybrid retrieval, and reranking
• Implement and scale vector search infrastructure (Pinecone, Weaviate, pgvector, or Elasticsearch) for semantic retrieval over large document corpora
• Develop agentic workflows with multi-step tool calling, function orchestration, and guardrails for safety and reliability
• Fine-tune open-weight and proprietary models using LoRA/QLoRA and instruction-tuning pipelines on domain-specific datasets
• Build systematic LLM evaluation frameworks: golden datasets, LLM-as-judge pipelines, regression suites wired into CI/CD
• Optimize inference performance and cost - quantization, batching, caching, and model routing across OpenAI, Anthropic, and self-hosted endpoints
• Deploy and monitor LLM services in production using observability tooling for latency, token cost, hallucination rates, and drift
What We Are Looking For
• 3–6 years of software or ML engineering experience, with at least 1–2 years building LLM/GenAI systems in production
• Strong Python engineering skills; comfort with async programming, REST/streaming APIs, and production code quality standards
• Hands-on experience with at least one LLM orchestration framework (LangChain, LlamaIndex, or equivalent) and one vector database
• Practical understanding of transformer architectures, embedding models, and retrieval-augmented generation patterns
• Experience deploying and operating services on AWS, GCP, or Azure, including containerized workloads (Docker, Kubernetes)
• MS in Computer Science, related field, or equivalent practical experience
• Bonus: experience with vLLM/TGI inference optimization, RLHF or preference tuning, Spark/PySpark data pipelines, or contributions to open-source LLM tooling