Senior AI Engineer / GenAI Solutions Architect

Job Description

Senior AI Engineer / GenAI Solutions Architect

NYC, NY

12+ Months

Role Overview

The ideal candidate combines strong software engineering skills with deep understanding of LLMs, RAG, agents, context management, evaluation, and scalability.

Key Responsibilities

Solution Design & Architecture

\n
• Design end-to-end AI, RAG, and agentic solutions for enterprise use cases.
\n
• Evaluate architectural trade-offs and select appropriate patterns, models, and platforms.
\n
• Create and defend Architecture Decision Records (ADRs) and technical designs.
\n
• Identify risks, failure modes, scalability concerns, and optimisation opportunities.
\n

AI Engineering & Development

\n
• Build production-grade AI applications using LLMs, agents, workflows, and retrieval systems.
\n
• Develop and integrate tools, APIs, vector databases, and knowledge systems.
\n
• Implement memory, context management, guardrails, evaluation, and observability capabilities.
\n
• Leverage AI-assisted coding tools (Claude Code, Cursor, GitHub Copilot, etc.) while maintaining engineering ownership of the solution.
\n

Production Readiness

\n
• Improve consistency, reliability, and performance of AI systems.
\n
• Troubleshoot issues such as hallucinations, context bloat, latency, cost overruns, and output variability.
\n
• Design monitoring, testing, evaluation, and governance frameworks for production systems.
\n
• Optimize inference, retrieval, caching, and overall system performance.
\n

Collaboration

\n
• Work with product, architecture, data, and platform teams to define and deliver solutions.
\n
• Translate business requirements into scalable technical architectures.
\n
• Contribute to engineering standards, best practices, and reusable AI assets.
\n

Required Skills & Experience

Core AI & LLM Engineering

\n
• Hands-on experience building GenAI, RAG, and agentic applications.
\n
• Strong understanding of LLM architectures, prompting, model selection, and evaluation.
\n
• Experience with multi-agent systems, tool calling, MCP, workflow orchestration, or similar patterns.
\n
• Understanding of fine-tuning, embeddings, vector search, and retrieval architectures.
\n

Architecture & System Thinking

\n
• Ability to justify technology choices and architectural decisions.
\n
• Experience designing solutions for enterprise-scale workloads and large data sets.
\n
• Strong understanding of scalability, reliability, cost, performance, and maintainability trade-offs.
\n
• Familiarity with Architecture Decision Records (ADR) and solution documentation.
\n

Context & Memory Management

\n
• Understanding of:
\n
• Context management strategies
\n
• Context compression and summarization
\n
• Short-term and long-term memory patterns
\n
• Retrieval optimisation
\n
• Token and prompt efficiency
\n

Engineering & Coding

\n
• Strong programming skills in Python and modern software engineering practices.
\n
• Experience with version control, testing, CI/CD, code reviews, and SDLC processes.
\n
• Ability to read, review, optimise, and troubleshoot AI-generated code.
\n

Optimization & Production Operations

\n
• Understanding of:
\n
• KV Cache
\n
• Prompt caching
\n
• Response caching
\n
• Guardrails
\n
• Evaluation frameworks
\n
• Monitoring and observability
\n
• Performance optimisation techniques
\n

What Success Looks Like

A successful candidate can confidently answer questions such as:

\n
• When should RAG be used instead of fine-tuning?
\n
• Why choose a multi-agent architecture versus a workflow?
\n
• How would you manage a million-token context?
\n
• Why is an AI agent producing inconsistent outputs?
\n
• Where should guardrails be implemented?
\n
• Would KV caching help in this scenario? Why and where?
\n
• What happens when retrieval quality degrades at enterprise scale?
\n

Ideal Candidate Profile

We are looking for someone who is:

✅ A strong engineer first

✅ Experienced with production AI systems

✅ Comfortable with ambiguity and problem solving

✅ Able to reason about architectural trade-offs

✅ Curious, pragmatic, and continuously learning

✅ Capable of designing solutions, not just building demos