AI Engineer

JD:
1. Experience building agent-based architectures, such as tool-using agents, workflow agents, multi-agent collaboration systems, and autonomous/goal-seeking pipelines using frameworks like LangChain, LlamaIndex, Haystack, or custom orchestration layers.

2. Experience designing, architecting, and deploying end-to-end AI systems, with a deep focus on building agentic workflows, autonomous orchestration patterns, and multi-step reasoning pipelines that integrate reliably with production systems.

3. Proficiency with large language models (LLMs) and modern AI frameworks, including hands-on experience with model APIs, fine-tuning techniques, retrieval-augmented generation (RAG), function calling, tool use, and structured output enforcement.

4. Advanced prompt engineering capabilities, including development of reusable prompt templates, prompt-chaining patterns, system instruction design, context management, and optimization of prompts for accuracy, determinism, and cost efficiency.

5. Deep understanding of retrieval and knowledge systems, including vector databases (e.g., Pinecone, Weaviate, Chroma, Vertex Matching Engine), embedding model selection, chunking strategies, and indexing techniques for low-latency, high-recall retrieval.

6. Designing AI workflows that integrate with business systems, including function-calling APIs, microservices, internal tools, CRMs, productivity suites, cloud tasks, and event-driven architectures.

7. Strong grounding in model evaluation and safety, including development of automated evaluation harnesses, guardrail systems, toxic output mitigation, hallucination reduction, and structured monitoring for output correctness and drift.

8. Experience with production-grade AI infrastructure, including serverless orchestration, containerized deployments, GPU/accelerator environments, and CI/CD workflows tailored for LLM-based systems.

9. Python engineering skills, including writing clean, testable, maintainable code, and building robust libraries, utilities, and agent behaviors that support long-running or complex AI workflows.

10. Hands-on experience with cloud-native AI tooling, such as: a. AWS: Bedrock, Sagemaker, Lambda, Step Functions, DynamoDB, S3, API Gateway b. GCP: Vertex AI, Cloud Functions, Cloud Run, Pub/Sub, BigQuery, GKE with strong understanding of security, identity, networking, and cross-service orchestration for AI-heavy systems.

11. Experience building monitoring and observability layers for AI systems, including prompt/response logging, latency tracking, quality scoring, conversation replay, and anomaly detection for agent performance