Job Description

Job Summary

Design, build, and operate enterprise-scale AI automation solutions—spanning RAG pipelines, LLM agents, document understanding, and decisioning—with a focus on quality, reliability, performance, and measurable outcomes.

6–10+ years total; 2–4+ years applied AI/NLP; multiple production launches.

Key Responsibilities

• Design RAG architectures including chunking, embedding strategies, retrieval, reranking, grounding, and evaluation.
• Build tool-enabled LLM agents with deterministic fallbacks, retry flows, and defensive error-handling patterns.
• Implement document processing workflows (OCR, layout extraction, entity extraction, tables) using industry-grade frameworks.
• Develop and operate inference APIs and integrate AI pipelines with event-driven and streaming platforms.
• Build evaluation suites leveraging golden datasets, LLM-as-judge frameworks, and rubric scoring.
• Add deep telemetry for hallucination detection, accuracy monitoring, latency profiling, and cost management.
• Optimize prompts, retrieval strategies, and inference paths for reliability, speed, and cost efficiency.
• Implement robust security controls: input validation, prompt-injection hardening, PII scrubbing/redaction.
• Collaborate with domain SMEs to design workflows, state machines, and business logic.

Develop reusable automation components, internal SDKs, and best-practice templates.

Skill Requirements

• Strong Python, embeddings, vector search, and RAG/agent tooling (LangChain, Semantic Kernel, DSPy).
• Experience with at least one enterprise LLM provider (Azure OpenAI, OpenAI, Anthropic, etc.).
• Familiarity with OCR/NLP frameworks, model registries, CI/CD, Docker/K8s, and observability (OpenTelemetry).
• Knowledge of secure AI practices, cost optimization, and evaluation techniques.

Other Requirements

null