AI Engineer

INFOSYS NOVA HOLDINGS LLC•Fort Worth, TX

About The Position

We are seeking a highly skilled Senior Gen AI / Agentic AI Engineer to design, build, and deploy enterprise-grade Generative AI and Agentic AI platforms. The role requires strong hands-on experience across LLMs, RAG, Graph RAG, multi-agent orchestration, vector databases, MCP setup, full-stack application development, cloud-native deployments, observability, and data engineering pipelines. The ideal candidate should be capable of building scalable AI platforms from end to end, including data ingestion, embedding pipelines, retrieval systems, agent workflows, model serving, API layers, UI/UX integration, monitoring, security, and production deployment across multi-cloud environments.

Requirements

Strong hands-on experience in Generative AI, LLMs, Agentic AI, RAG, Graph RAG, and prompt engineering.
Experience building multi-agent AI systems using LangChain, LangGraph, CrewAI, AutoGen, Semantic Kernel, or similar frameworks.
Strong understanding of LLM orchestration, tool calling, function calling, agent memory, planning, reasoning, task decomposition, and workflow automation.
Hands-on experience with RAG architecture, including chunking strategies, embeddings, vector search, hybrid search, reranking, metadata filtering, and evaluation.
Experience with vector databases such as Pinecone, Weaviate, Milvus, FAISS, Chroma, Azure AI Search, OpenSearch, Vertex AI Vector Search, or pgvector.
Knowledge of Graph RAG / knowledge graph solutions using Neo4j, Neptune, TigerGraph, RDF, SPARQL, Cypher, or graph-based retrieval patterns.
Strong backend development experience with Python, FastAPI, Flask, Node.js, Java, REST APIs, GraphQL, microservices, and event-driven systems.
Full-stack development experience with React, Angular, Next.js, TypeScript, JavaScript, HTML, CSS, and API integration.
Experience with model serving and inference optimization using vLLM, Hugging Face, TGI, Triton, TensorRT-LLM, Ray Serve, or similar platforms.
Strong cloud experience across AWS, Azure, and/or GCP.
Hands-on experience with Docker, Kubernetes, Helm, Terraform, CI/CD pipelines, and cloud-native deployments.
Strong data engineering skills using SQL, Python, Spark/PySpark, Databricks, Airflow, Kafka, and cloud data platforms.
Experience implementing observability, monitoring, logging, tracing, and evaluation for AI/ML/LLM applications.
Strong understanding of LLMOps/MLOps, model lifecycle management, prompt lifecycle, evaluation metrics, and production support.
Experience with security, governance, responsible AI, guardrails, prompt injection protection, and enterprise compliance controls.

Nice To Haves

Hands-on experience setting up MCP server/client architecture for enterprise agentic AI platforms.
Experience with A2A / agent-to-agent communication, multi-agent collaboration, and tool registry patterns.
Experience with fine-tuning, LoRA, QLoRA, PEFT, RLHF, RLAIF, or domain-specific model adaptation.
Experience with document intelligence platforms, OCR, extraction pipelines, and intelligent search.
Experience with Databricks Mosaic AI, Azure AI Foundry, AWS Bedrock Agents, Google Vertex AI Agent Builder, or similar enterprise AI platforms.
Experience with semantic caching, token optimization, prompt compression, and cost optimization.
Familiarity with Neo4j, AWS Neptune, OpenSearch, Elasticsearch, Redis, MongoDB, PostgreSQL, Snowflake, BigQuery.
Experience in regulated industries such as banking, financial services, healthcare, or insurance.

Responsibilities

Design and develop enterprise Gen AI and Agentic AI applications using LLMs, RAG, Graph RAG, multi-agent workflows, and tool-augmented reasoning.
Build scalable RAG pipelines including document ingestion, chunking, embedding generation, metadata enrichment, hybrid search, reranking, retrieval optimization, and response grounding.
Implement Graph RAG solutions by integrating knowledge graphs, entity extraction, relationship mapping, graph traversal, and contextual retrieval.
Develop multi-agent systems using frameworks such as LangChain, LangGraph, CrewAI, AutoGen, Semantic Kernel, LlamaIndex, and custom orchestration patterns.
Set up and integrate MCP servers and clients to enable tool connectivity, enterprise system integration, agent-to-tool communication, and reusable AI workflows.
Build and manage vector database solutions using Pinecone, Weaviate, Milvus, FAISS, Chroma, OpenSearch Vector Engine, Azure AI Search, Vertex AI Vector Search, or pgvector.
Deploy and optimize LLM / VLLM inference stacks using vLLM, Hugging Face Transformers, TensorRT-LLM, TGI, Ollama, llama.cpp, Ray Serve, or Triton Inference Server.
Integrate commercial and open-source LLMs such as GPT, Claude, Gemini, Llama, Mistral, Mixtral, Falcon, Cohere, DeepSeek, and domain-specific fine-tuned models.
Develop secure and scalable backend services using Python, FastAPI, Flask, Node.js, Java/Spring Boot, or similar API frameworks.
Build full-stack applications with frontend technologies such as React, Angular, Next.js, TypeScript, JavaScript, HTML, CSS, and integrate AI workflows into user-facing interfaces.
Create intuitive UI/UX experiences for AI chatbots, agent workbenches, document intelligence platforms, prompt playgrounds, feedback loops, approval workflows, and human-in-the-loop systems.
Implement data engineering pipelines using Spark, PySpark, Databricks, Airflow, Kafka, Snowflake, BigQuery, Redshift, SQL, NoSQL, and cloud-native data services.
Build ingestion pipelines for structured, semi-structured, and unstructured data including PDFs, Word documents, emails, images, logs, databases, APIs, and enterprise repositories.
Deploy AI workloads on AWS, Azure, and GCP, using services such as Bedrock, SageMaker, Azure OpenAI, Azure AI Search, Azure ML, Vertex AI, BigQuery, GKE, AKS, EKS, Lambda, and Cloud Functions.
Implement cloud-native architecture using Docker, Kubernetes, Helm, Terraform, CI/CD, GitHub Actions, GitLab, Jenkins, and infrastructure-as-code practices.
Establish strong monitoring and observability for Gen AI applications, including prompt/response tracing, token usage, latency, hallucination tracking, retrieval quality, cost monitoring, model drift, and agent execution traces.
Use tools such as LangSmith, Arize Phoenix, W&B Weave, MLflow, Evidently AI, Prometheus, Grafana, OpenTelemetry, Splunk, Datadog, ELK, and cloud-native logging platforms.
Implement LLMOps / MLOps practices including model registry, prompt versioning, evaluation pipelines, A/B testing, guardrails, feedback capture, safety checks, and automated deployment.
Apply security and governance controls including PII detection, data masking, access control, RBAC, IAM, encryption, audit logging, policy enforcement, prompt injection prevention, and responsible AI guardrails.
Collaborate with product owners, architects, data scientists, engineers, UX teams, security teams, and business stakeholders to deliver production-grade AI solutions.
Optimize Gen AI applications for accuracy, latency, scalability, reliability, cost, and user experience.