Senior AI Engineer (LLM & Applied ML)

Softadom; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

Softadom is seeking an applied AI engineer to embed language models into our production web software, turning demos into scalable, user‑facing features. You will ship LLM‑backed capabilities across backend and frontend surfaces, tackling latency, cost, and guardrails as engineering problems with measurable outcomes.

You'll define evaluation metrics, instrument pipelines with structured logging, and collaborate with security and platform teams on data handling, privacy, and reliability to ensure

Qualifications

• Strong Python skills with hands-on shipping LLM-backed features.
• Experience across a web stack (backend framework + modern frontend framework).
• Familiarity with vector stores and retrieval patterns.
• Hands-on prompt engineering treated as measurable discipline.
• Collaborating with data privacy and reliability concerns with cross-functional teams.

Responsibilities

• Embed language models into backend and frontend product surfaces.
• Define guardrails, metrics, and quality benchmarks beyond anecdotal checks.
• Optimize latency, throughput, and provider spend as usage scales.
• Instrument pipelines with structured logging for traceability.
• Collaborate with security and platform teams on data handling and failure modes.

Skills

Python

Web stack

Frontend framework

Backend framework

Vector stores

Retrieval patterns

Background jobs

Prompt engineering

Cross-functional collaboration

Tools

Vector DBs

Model deployment tools

Job description

Role overview

An applied AI engineering role focused on embedding language models into production business software rather than demo prototypes. Work centers on integrating models into existing web products while treating prompts, guardrails, latency, and cost as engineering problems with measurable outcomes.

Responsibilities

• Integrate large language models into backend and front-end product surfaces, including intelligent search, document processing, workflow automation, and agent-style assistant experiences inside customer-facing portals.
• Define and enforce guardrails, evaluation metrics, and quality benchmarks that go beyond anecdotal checks.
• Optimize for latency, throughput, and provider spend as usage scales across customer accounts.
• Instrument pipelines with structured logging so every model call can be traced, audited, and reproduced.
• Collaborate with security and platform teams on PII boundaries, data handling policies, and documented failure modes.

Requirements

• Strong Python skills with hands‑on experience shipping LLM‑backed features, not just prototyping them.
• Comfort working across a typical web stack (a backend framework plus a modern front‑end framework) and integrating hosted model providers or comparable alternatives.
• Familiarity with vector stores, retrieval patterns, background job systems, and the operational side of model calls.
• Practical prompt engineering experience treated as a measurable discipline rather than intuition.
• Track record of collaborating with cross‑functional partners on data, privacy, and reliability concerns.

Nice to have

• Prior work on AI features that real users rely on, with evidence of iteration driven by observed behavior.