Full Stack Engineer, AI Systems

Salt Digital Recruitment; United States; On-site; USD 110,000 - 180,000; Full time

Job summary

Salt Digital Recruitment is seeking a Strong Full Stack Engineer to build AI-native features spanning frontend, backend and AI integrations in a fast-moving environment. You will design agent-driven workflows, enable tool use, memory, and persistent context to deliver reliable, multi-step tasks with low latency.

You will work with Next.js, Python, Node.js, PyTorch, and Kubernetes to productionize AI experiences and ensure robust failure handling and observability across systems.

Qualifications

• Strong full stack engineering experience across frontend and backend.
• Solid understanding of system design, API architecture and production software engineering.
• Hands-on experience building with LLMs, RAG systems, AI agents or AI-powered applications.
• Experience integrating AI models and external services into production systems.
• Understanding of latency, reliability and failure handling in AI apps.

Responsibilities

• Build and ship end-to-end product features across frontend, backend and AI integrations.
• Design agent workflows capable of planning, tool use, multi-step execution, failure handling and recovery.
• Integrate LLMs, memory systems and external tools into reliable production applications.
• Build real-time AI experiences with streaming and tight latency constraints.
• Develop robust fallback and recovery mechanisms for model and tool failures.
• Improve system reliability, observability and workflow success rates.
• Collaborate with ML, backend, product and engineering teams to deliver features end-to-end.
• Evaluate real-world usage and failure modes to improve AI experiences.
• Develop reusable patterns for integrating AI models, memory and tools into scalable systems.
• Balance speed of delivery with quality, scalability and reliability.

Skills

Full stack

System design

API architecture

LLMs & RAG

Tool integration

Latency & reliability

Ownership

Ambiguity handling

Independent work

Cross-functional

Tools

Next.js

Python

Node.js

PyTorch

Kubernetes

Docker

Job description

About the Opportunity

We are partnering with a fast-growing technology company building a new generation of AI-native applications designed to make everyday tasks, communication, organization and workflows more intelligent and intuitive. The team is developing proactive AI experiences that require minimal prompting, with a strong focus on persistent context, reliable long-running workflows and successful real-world task completion. They are looking for a Full Stack Engineer - AI Systems to build the product layer that transforms advanced AI capabilities into intuitive, reliable and production-ready experiences.

The Role

As a Full Stack Engineer, you will work across frontend, backend and AI systems, taking ownership of features from initial concept through to production. A major focus of the role will be building agentic AI workflows that can plan, use tools, maintain context, recover from failures and reliably complete multi-step tasks. You'll work at the intersection of product engineering and applied AI, helping create experiences that move beyond traditional chat interfaces toward persistent, goal-driven AI applications.

What You'll Be Doing

• Build and ship end-to-end product features across frontend, backend and AI integrations.
• Design agent workflows capable of planning, tool use, multi-step execution, failure handling and recovery.
• Integrate LLMs, memory systems and external tools into reliable production applications.
• Build real-time AI experiences incorporating streaming and partial results while operating within tight latency requirements.
• Develop robust fallback and recovery mechanisms for model and tool failures.
• Improve system reliability, observability and overall workflow success rates.
• Work closely with machine learning, backend, product and engineering teams to deliver features end-to-end.
• Evaluate real-world usage and failure modes to continuously improve AI-driven experiences.
• Develop reusable patterns and abstractions for integrating AI models, memory and external tools into scalable product systems.
• Balance speed of delivery with engineering quality, scalability and reliability.

What We're Looking For

• Strong full stack engineering experience across both frontend and backend development.
• Solid understanding of system design, API architecture and production software engineering.
• Hands-on experience building with LLMs, RAG systems, AI agents or AI-powered applications.
• Experience integrating AI models and external services into production systems.
• Understanding of the challenges associated with latency, reliability and failure handling within AI applications.
• Strong ownership mentality with the ability to take features from idea through to production.
• Ability to work through ambiguity and make pragmatic engineering decisions.
• Comfortable working independently within a fast-moving environment where requirements and technologies continue to evolve.
• Strong collaboration skills and the ability to work across product, ML and engineering functions.

Technology Environment

• Next.js
• Python
• Node.js
• PyTorch
• Commercial and open-source LLMs
• SQL and NoSQL databases
• Kubernetes
• Docker
• Retrieval-Augmented Generation (RAG)
• Agentic AI and tool-calling architectures
• AI memory and context systems

What Success Looks Like

You will own and ship AI-native product features that move beyond simple conversational experiences into persistent, goal-driven workflows. You will design and deploy AI agents capable of reliably completing multi-step tasks across tools and sessions, while improving latency, responsiveness and output quality. Through monitoring, evaluation and continuous iteration, you will improve the reliability and success rate of AI-powered workflows and establish scalable patterns for integrating LLMs, memory and external tools into production systems. Ultimately, your work will help create AI experiences that feel proactive, consistent and dependable rather than simply reactive.