our Role

• Integrate LLMs (Llama, OpenAI, Claude, Mistral, Gemini) and embedding models into applications.

• Implement agentic workflows using LangGraph, AgentCore, Google SDK, applying patterns such as Planner-Executor, Reflection Loops, and Multi-Agent Orchestration.

• Build and optimize data ingestion and retrieval pipelines including preprocessing, chunking, enrichment, embeddings, and vector indexing.

• Integrate coarse-grained controls (data masking, compliance filters, enterprise-wide security checks).

• Prototype in Jupyter Notebooks and transition solutions to production.

• Implement LLMOps practices:

• Instrument pipelines for metrics (latency, cost, hallucination rate, accuracy, compliance violations).

• Build dashboards, alerts, and feedback loops.

• Automate evaluations (A/B testing, prompt regression checks, continuous accuracy testing).

• Collaborate with Architect to integrate knowledge graphs where applicable.

• Deploy and maintain solutions on AWS, GCP, or Azure leveraging GPU/compute environments.

Your Profile

• Strong programming in Python; working knowledge of JavaScript.

• Experience developing on cloud platforms (AWS, GCP, Azure) with GPU-backed compute.

• Proficiency with Jupyter Notebooks for experimentation.

• Hands-on with LLMs, embeddings, and vector search pipelines.

• Experience implementing agentic workflows with security guardrails.

• Familiarity with knowledge graphs is a plus.

• Proven experience in LLMOps implementation (logging, monitoring, evaluation pipelines, observability tools).

• Strong debugging, problem-solving, and execution focus.