AI Architect - GenAI / ML (New York City Metropolitan Area)

AI Architect – GenAI / ML

Role Overview

Photon is looking for a hands-on AI Architect – GenAI / ML to support large-scale AI-led transformation initiatives within the financial services industry.

This role will be responsible for defining and implementing solution architectures across Generative AI, Machine Learning, agentic AI, enterprise data, APIs, and application platforms .

The ideal candidate will combine strong AI architecture experience with deep hands-on expertise in Python, ML frameworks, LLM-based applications, orchestration frameworks, and production AI engineering .

Key Responsibilities

• Define the end-to-end architecture for GenAI and ML solutions , from data ingestion and feature engineering through model execution, orchestration, integration, and production deployment.
• Design enterprise-grade solutions leveraging LLMs, traditional ML models, retrieval-augmented generation, agents, and AI orchestration frameworks .
• Build and validate prototypes and reference implementations using Python .
• Architect RAG solutions , including document ingestion, chunking, embeddings, retrieval, reranking, prompt construction, and response generation.
• Design agentic AI architectures supporting tool usage, workflow orchestration, reasoning, memory, and multi-agent interactions.
• Define patterns for integrating AI capabilities with enterprise applications, APIs, data platforms, event streams, and legacy systems .
• Partner with Data Scientists, ML Engineers, Data Engineers, application teams, and business stakeholders to translate business use cases into scalable AI solutions.
• Define reusable AI architecture patterns, frameworks, and components that can be applied across multiple enterprise use cases.
• Establish patterns for prompt management, model abstraction, model routing, model versioning, and evaluation .
• Design architectures for model inference, feature pipelines, model serving, and real-time or batch scoring.
• Implement and guide development of Python-based AI services, APIs, pipelines, and orchestration components .
• Ensure AI solutions meet requirements for security, privacy, scalability, reliability, performance, explainability, governance, and auditability .
• Evaluate open-source and commercial AI frameworks and determine appropriate technologies based on enterprise requirements.
• Define technical approaches for model monitoring, hallucination detection, evaluation, guardrails, observability, and human-in-the-loop controls .
• Lead architecture and code reviews and provide technical guidance through implementation and production rollout.

Required Experience

• Strong experience as an AI Architect / ML Architect / GenAI Architect / Lead AI Engineer within large-scale enterprise environments.
• Deep hands-on programming experience with Python .
• Strong understanding of Machine Learning and Deep Learning fundamentals .
• Hands-on experience with frameworks such as PyTorch, TensorFlow, scikit-learn, or equivalent .
• Strong experience building LLM-based applications and GenAI solutions .
• Experience working with LLM orchestration frameworks such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, or equivalent frameworks .
• Strong experience with RAG architectures, embeddings, vector search, semantic retrieval, and reranking .
• Experience designing and building agentic AI solutions with tool calling, workflow orchestration, memory, and multi-step reasoning.
• Strong understanding of prompt engineering, prompt versioning, prompt evaluation, and structured outputs .
• Experience building REST APIs and backend AI services using frameworks such as FastAPI, Flask, or equivalent .
• Strong understanding of model lifecycle concepts including training, fine-tuning, inference, evaluation, monitoring, and versioning .
• Experience integrating AI solutions with enterprise data sources, APIs, databases, messaging systems, and application platforms .
• Understanding of data pipelines, feature engineering, data quality, and model input validation.
• Experience building scalable and production-ready AI systems rather than only proof-of-concepts.

GenAI / LLM Expertise

• Experience designing solutions using large language models and foundation models .
• Strong understanding of:
• Prompt engineering
• RAG architectures
• Embeddings and vector retrieval
• Tool/function calling
• Agentic workflows
• Structured outputs
• Context management
• Model routing
• Guardrails
• Evaluation frameworks
• Hallucination mitigation
• Ability to evaluate when to use RAG, fine-tuning, traditional ML, rules-based logic, or agentic approaches depending on the business problem.
• Experience designing enterprise AI systems that can work across multiple models without creating tight vendor dependency.

Machine Learning Expertise

• Strong understanding of supervised and unsupervised learning, classification, regression, clustering, anomaly detection, and recommendation approaches .
• Experience with feature engineering, feature selection, model training, model validation, and inference.
• Understanding of ML evaluation metrics and model performance analysis.
• Experience integrating ML models into enterprise applications and operational workflows.
• Familiarity with ML pipelines, model registries, experiment tracking, and model monitoring.

AI Governance & Production Readiness

• Define controls for model and prompt versioning, evaluation, observability, explainability, and traceability .
• Design AI systems with appropriate security, privacy, data protection, and access controls .
• Establish approaches for human review and escalation where AI outputs require oversight.
• Support implementation of automated evaluation and testing across GenAI and ML solutions.
• Ensure solutions can be monitored for quality degradation, model drift, hallucinations, latency, and reliability .

Preferred Experience

• Experience within Banking, Payments, Fraud, Cards, Financial Crime, Wealth, or other financial-services domains .
• Experience developing AI solutions using customer, transaction, payment, behavioral, or risk data .
• Experience building AI-enabled decisioning, fraud detection, investigation, or operational automation solutions.
• Experience working in complex on-premise or hybrid enterprise environments .
• Familiarity with distributed data technologies such as Spark and Kafka .
• Experience with enterprise AI platforms, model gateways, or centralized AI orchestration platforms.

Key Profile We Are Looking For

A hands-on AI Architect who can architect, code, prototype, and guide implementation .

The individual should be comfortable moving across:

Business Use Case → Data → ML / LLM → RAG / Agents → Python Services → APIs → Enterprise Integration → Evaluation → Production

The ideal candidate should be able to write Python, build a working AI prototype, troubleshoot model or RAG behavior, design the enterprise architecture, and guide engineering teams to productionize the solution .