3 days ago
Pune, IndiaStaff+
Responsibilities
- Architect and operationalize end-to-end autonomous agentic AI systems, including multi-agent orchestration, tool calling, reflection, planner-executor loops, and persistent state management.
- Design scalable LLM runtime infrastructure using retrieval, graph-based context grounding, vector search, reranking, semantic routing, evaluation, guardrail, and observability frameworks.
- Make architectural decisions involving frontier models, smaller language models, fine-tuning, LoRA, quantization, and self-hosted or managed inference endpoints.
- Write production code, develop framework primitives, and rapidly prototype and validate zero-to-one AI concepts.
- Build resilient, secure, fault-tolerant, and performant AI/ML systems, pipelines, and applications for high-throughput real-time environments.
- Integrate Model Context Protocol and agent-to-agent communication protocols.
- Partner with AI, product, and VP-level stakeholders on the AI technology roadmap and emerging AI capabilities.
- Mentor engineers, provide technical leadership, and contribute to open-source projects, technical papers, conferences, or industry forums.
Requirements
- 12+ years of software development experience, with the role description also requesting 15+ years of hands-on development experience.
- 5+ years actively building and deploying generative AI and agentic AI applications in production.
- 8+ years writing production-grade code in Python or similar languages, with expertise in scalable software design, APIs, and microservices.
- Demonstrated ability to independently architect, code, and deploy custom AI tools, autonomous agents, and bots.
- Deep proficiency with LangChain, LlamaIndex, Auto-GPT, CrewAI, and Model Context Protocol.
- Practical experience with GPT-4, Gemini, Claude, fine-tuning, prompt engineering, and agentic tool-calling workflows.
- Experience training, tuning, and running models with PyTorch, TensorFlow, or Scikit-learn.
- Familiarity with vLLM, TensorRT-LLM, TGI, NVIDIA GPUs, and TPUs.
- Experience building low-latency vector search indexes, in-memory caching layers, streaming pipelines, and NoSQL data stores.
- Exposure to AWS, Azure, or GCP and containerized deployment with Kubernetes and Docker.
- A bachelor’s degree or equivalent practical experience is required; a master’s or Ph.D. in computer science, AI, ML, or a related quantitative field is preferred.
- Experience in contact center, telecommunications, or customer experience platforms is preferred.
- Excellent technical communication, ownership, mentoring, and senior individual-contributor leadership skills are required.
Benefits
- Applicants must be currently authorized to work in the United States without current or future visa sponsorship.
- Avaya is an equal opportunity employer and a U.S. Federal Contractor.
Tech Stack
Apache FlinkApache KafkaAWSAzureDockerGoogle Cloud PlatformKubernetesPythonPyTorchscikit-learnTensorFlow
Categories
About Avaya
Avaya builds enterprise communications and contact-center software delivered as UCaaS, CCaaS, and CPaaS, along with services and devices, for large organizations and government agencies. Its portfolio—including the Avaya Infinity platform—supports cloud, hybrid, and on-prem deployments and modernization of legacy voice systems under subscription and support contracts. Founded in 2000 as a Lucent spinoff and now privately held under Silver Lake, Avaya is headquartered in Morristown, New Jersey and serves customers worldwide.