Software Engineer

Agent Architecture: Design and implement scalable AI agents, autonomous multi-agent systems, and tool-use pipelines using LangChain/LangGraph or Google ADK.

Backend Development: Build robust, high-performance REST/gRPC APIs and microservices in Python (FastAPI / Flask / AsyncIO) to integrate LLMs into production environments.

RAG & Vector Search: Implement Advanced Retrieval-Augmented Generation (RAG) pipelines using vector databases (Pinecone, Qdrant, Chroma, Weaviate, or Pgvector) with semantic search and re-ranking.

LLM Integration & Optimization: Integrate state-of-the-art foundation models (Gemini, OpenAI, Anthropic, open-source models via Ollama/vLLM), optimizing for latency, throughput, token usage, and cost.

Tooling & Orchestration: Connect agents to external APIs, databases, function calls, and enterprise tools (using MCP/Model Context Protocol or custom connectors).

Evaluation & Testing: Establish deterministic testing, evaluation harnesses, guardrails (e.g., NeMo Guardrails, LangSmith, Google ADK Evaluation), and monitoring for hallucination, safety, and compliance.

Cloud & DevOps: Containerize and deploy LLM workloads on cloud platforms (GCP / AWS / Azure) using Docker, Kubernetes, or managed agent runtime environments.