Backend / AI Engineer

LastRound AI; United States; Remote; USD 120,000 - 170,000; Full time

Job summary

LastRound AI is seeking an engineer to build the real-time interview assistant AI engine, turning live questions into fast, context-aware answers. You’ll work across speech recognition, prompts, retrieval, and orchestration layers to deliver a streaming, low-latency experience.

Responsibilities include developing speech-to-text pipelines, designing prompts and retrieval logic, and ensuring performance and reliability through data-driven improvements.

Qualifications

• Proficiency in Python for AI systems and data processing.
• Hands-on experience with applied AI, especially large language models.
• Familiarity with speech-to-text, retrieval, or prompt engineering concepts.
• Discipline around performance, latency, and production reliability.
• Comfortable measuring impact and iterating based on data.

Responsibilities

• Develop and tune speech-to-text pipelines for live interview audio.
• Engineer prompts and retrieval logic over resumes and candidate notes.
• Orchestrate large language models into a low-latency, streaming response flow.
• Measure the quality, speed, and reliability of every model change.
• Tune systems relentlessly based on observed performance and user signals.

Skills

Python

Applied AI/LLMs

Speech-to-Text

Retrieval

Prompt Engineering

Performance Monitoring

Job description

Role overview

Build the AI engine behind a real-time interview assistance product, turning live spoken questions into instant, context-aware answers. The work sits at the technical heart of the product, where latency, reliability, and answer quality all matter at once. Expect to spend time across speech recognition, prompt design, retrieval, and the orchestration layer that ties large language models into a fast streaming experience.

Responsibilities

• Develop and tune speech-to-text pipelines for live interview audio
• Engineer prompts and retrieval logic over resumes and candidate notes
• Orchestrate large language models into a low-latency, streaming response flow
• Measure the quality, speed, and reliability of every model change
• Tune systems relentlessly based on observed performance and user signals

Requirements

• Strong proficiency in Python
• Hands-on experience with applied AI, particularly large language models
• Familiarity with speech-to-text, retrieval, or prompt engineering concepts
• Discipline around performance, latency, and production reliability
• Comfort measuring impact and iterating based on data

Nice to have

• Experience with real-time or streaming AI systems
• Background optimizing model inference cost and speed
• Curiosity about evaluation pipelines for AI-generated answers