AI Software Engineer

NearTech Search; United States; Remote; GBP 90,000 - 130,000; Full time

Job summary

NearTech Search is seeking an AI Software Engineer to build and improve data-processing systems for enterprise clients. You will own the full AI stack, integrate models from multiple providers, and deliver reliable, actionable insights.

Ideal candidates have 5+ years in software with production LLM experience, strong Python, and hands-on work with vector embeddings and cloud platforms. Fully remote role with a lean core team and ownership of end-to-end projects.

Qualifications

• 5+ years software development with production LLM experience.
• Strong Python (3.10+) and async cloud functions.
• Experience with multi-model AI pipelines and vector embeddings.

Responsibilities

• Own full AI software stack: data processing, routing, model integration, and output assembly.
• Take features from concept to production with monitoring of real-world performance.
• Improve accuracy across data quality, prompts, and model selection.
• Integrate third-party services with reliable failover.
• Create test suites and metrics to validate changes.
• Configure client-specific settings without maintenance overhead.
• Gather end-user feedback on usage of your code.

Skills

Python

LLM systems

Async workflows

Testing & metrics

Enterprise customer communication

Technical leadership

Tools

Docker

Terraform

Azure

AWS

GCP

Node.js

Job description

A well-funded startup is looking for an AI Software Engineer to build and improve systems that process complex data inputs for enterprise clients, delivering reliable, actionable insights. You'll work with a lean, experienced team who care about getting things right.

What You'd Do

• Own the full technical stack as an AI Software Engineer: data processing, routing logic, model integration, external service connections, and output assembly
• Take features from concept to production and monitor real-world performance
• Identify accuracy gaps and solve them across the stack (data quality, prompts, model selection, or replacing ML with logic where it makes sense)
• Connect third-party services reliably and fail gracefully when they don't work
• Create test suites and metrics that validate your changes
• Handle client-specific configurations without creating maintenance nightmares
• Get direct feedback from end users about how your work gets used

What They're Looking For

• 5+ years building software; substantial recent experience with production LLM systems
• Strong Python skills (3.10+); comfortable building resilient, asynchronous workflows using cloud function frameworks
• Experience building and deploying agentic systems using LLM abstraction layers, with hands‑on work across multiple model providers (Gemini, Claude, etc.)
• Proficiency with vector embeddings and retrieval optimization for AI pipelines
• Experience containerising applications (Docker) and managing infrastructure-as-code (Terraform or similar)
• Comfortable working across cloud platforms (Azure, AWS, GCP) and integrating with services/APIs written in other languages (particularly Node.js)
• You have opinions about what should be handled by models vs. what belongs in code
• You've written test data, defined success metrics, and seen them catch real issues
• You validate your own work critically before others doYou investigate what data the model is actually seeing before tweaking prompts
• You identify gaps and propose solutions without needing a task list
• Able to lead technical discussions with enterprise customers — explaining complex AI architecture with clarity and confidence
• Fully remote, small core team
• Focus on shipping incrementally and learning from real usage
• Individual ownership of projects from design through deployment
• Lightweight processes that stay out of your way
• Regular input from the people who depend on your code