• End-to-End Agent Orchestration

• • Design, build, and maintain scalable, high-performance microservices that orchestrate autonomous agent workflows: prompt planning, tool invocation, multi-step decision loops, and response generation.
  • Implement context management, memory retrieval, and streaming responses to support real-time conversational agents.
• LangGraph & Pipeline Tooling

• • Leverage LangGraph to define dynamic, graph-based LLM pipelines—authoring nodes, edges, and execution logic for complex agent behaviors (e.g., conditional branching, tool use, external API calls).
• MLOps & CI/CD for Agents

• • Develop robust infrastructure to automate model training, versioning, testing, and deployment.
  • Implement continuous integration, canary releases, blue-green deployments, and rollback strategies for both base LLMs and agent orchestration code.
• Cross-Functional Collaboration

• • Partner with Product Managers, Business Analysts, and AI Researchers to translate novel agent architectures into production-ready services.
  • Communicate design trade-offs, performance benchmarks, and roadmap items to both technical and non-technical stakeholders.

• Experience & Languages

• • 2+ years building backend services for ML/AI products (Chatbot, AI Agent, or Computer Vision).
  • Hands-on Python expertise in production.

• LLM & Conversational AI

• • Deep familiarity with at least one LLM framework (e.g., Hugging Face Transformers, OpenAI API).
  • Proven ability to architect prompt management, session/state persistence, fallback strategies, and safe-completion policies.
• LangGraph Expertise: Demonstrable track record designing and deploying LangGraph pipelines for multi-stage conversational or agent workflows.
• Systems & Scalability
• • Strong understanding of data structures, algorithms, network I/O, concurrency models, and distributed systems.
  • Experience with containerized, low-latency services using Kubernetes, Docker, and gRPC/REST.
• Software Engineering Best Practices
• • Competence in unit/integration testing, version control (Git), code reviews, and common design patterns

Preferred Skills

• High-Throughput Inference: Experience developing secure, low-latency, high-concurrency ML or web backend services—ideally supporting thousands of concurrent sessions.
• Model Serving Frameworks: Hands-on with two or more inference stacks (e.g., TorchScript/TorchServe, TensorRT, NVIDIA Triton Inference Server).
• AI Agent & Chatbot Development: Prior work building and deploying autonomous AI Agents or advanced Chatbot platforms in production.
• Container Orchestration: Deep familiarity with Docker, Kubernetes (including autoscaling and service meshes), and secure multi-tenant deployments.
• MLOps Platform: Experience with Kubeflow (Pipelines, Katib), MLflow, or similar for end-to-end ML lifecycle automation.