AI Engineer

INFOSYS NOVA HOLDINGS•New York, NY

About The Position

We are seeking an AI Engineer to build secure, production-ready AI services and agentic workflows. The role uses Python, FastAPI, Apache Airflow, retrieval-augmented generation (RAG), LLM APIs, LangGraph, LangChain, and MongoDB to deliver reliable capabilities across ingestion, retrieval, API deployment, evaluation, and operational support. The successful candidate will bring strong software-engineering discipline, practical AI judgment, and a clear understanding of security, responsible AI, and production operations. The AI Engineer designs and implements production-grade AI applications, copilots, and agentic workflows aligned with measurable business outcomes. The role builds maintainable Python services and secure, versioned FastAPI interfaces; designs RAG pipelines; integrates LLMs; and implements stateful workflows using LangGraph. It also involves operating Apache Airflow DAGs, designing MongoDB models and access patterns, and establishing testing, CI/CD, infrastructure configuration, observability, and operational runbooks. The role partners with product owners, data engineers, security, architecture, and subject-matter experts to translate requirements into safe technical designs. It also documents architecture decisions, model limitations, data lineage, controls, and support procedures. The employer is unable to sponsor or take over sponsorship of an employment visa at this time.

Requirements

  • Typically 4 or more years of software, data, or machine-learning engineering experience, including at least 1 year delivering LLM-enabled applications; equivalent depth demonstrated through shipped products is acceptable.
  • Demonstrable ownership of at least one AI solution beyond proof of concept, with production evidence such as monitoring, security controls, evaluation, support, and measurable outcomes.
  • Bachelor’s degree in computer science, engineering, mathematics, or a related discipline, or equivalent professional experience.
  • Strong skills in Python, FastAPI and APIs, Apache Airflow, RAG, LLMs, LangGraph, LangChain, and MongoDB.
  • Hands-on understanding of Model Context Protocol (MCP) and experience building or integrating MCP servers and MCP clients.
  • Strong software-engineering discipline and a clear understanding of security, responsible AI, and production operations.

Nice To Haves

  • For a senior appointment: evidence of technical design leadership, mentoring, architecture decisions, and cross-team delivery.
  • Experience with model gateways and providers such as Azure OpenAI, Anthropic, or AWS Bedrock.
  • Experience with vector-search platforms or extensions such as MongoDB Vector Search, Elasticsearch/OpenSearch, or Milvus.
  • Experience with MLOps/LLMOps tooling for prompt/version management, experiment tracking, evaluation, and tracing, such as Galileo.
  • Cloud-native delivery experience using RedHat OpenShift or equivalent infrastructure as code.
  • Knowledge-graph, multimodal, fine-tuning, synthetic-data, or advanced agent-evaluation experience.
  • Experience in regulated or control-focused environments with formal architecture, risk, privacy, and change-management processes.

Responsibilities

  • Design and implement production-grade AI applications, copilots, and agentic workflows aligned to measurable business outcomes.
  • Build maintainable Python services and expose capabilities through secure, versioned FastAPI interfaces.
  • Design RAG pipelines covering ingestion, chunking, metadata, embedding, retrieval, reranking, citation, and evaluation.
  • Integrate and invoke LLMs through provider SDKs or model gateways, applying structured outputs, tool/function calling, retries, rate limits, token controls, and fallbacks.
  • Use LangGraph to implement stateful, controllable workflows and LangChain selectively for model, tool, retriever, and prompt integrations.
  • Build or integrate MCP servers and clients, and expose enterprise capabilities through MCP tools, resources, and prompts with appropriate schemas and controls.
  • Create and operate Apache Airflow DAGs for scheduled ingestion, embedding refresh, evaluation, batch processing, and data quality controls.
  • Design MongoDB document models, indexes, and access patterns for application state, conversation history, metadata, and operational records.
  • Implement automated testing, CI/CD, infrastructure configuration, logging, tracing, monitoring, and incident-ready runbooks.
  • Partner with product owners, data engineers, security, architecture, and subject-matter experts to translate requirements into safe technical designs.
  • Document architecture decisions, model limitations, data lineage, controls, and support procedures.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service