Lead Machine Learning Engineer

Harnham; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Harnham in San Francisco seeks an experienced software engineer to define and scale AI systems across our design platform. You will own AI architecture, ship production-grade LLM features, and drive end-to-end pipelines from prompting to evaluation.

Join a fast-paced team building real-time AI experiences with context-aware code generation and intelligent suggestions. You’ll collaborate closely with product, design, and engineering to deliver high-impact features.

Qualifications

• 5+ years of software engineering experience.
• 2+ years working with LLM APIs and prompt engineering in production.
• Strong proficiency in TypeScript or Node.js and Python.
• Experience with LLM evaluation, benchmarking, and performance measurement.
• Hands-on experience with fine-tuning (LoRA, full fine-tuning, distillation).
• Experience building agent-based AI systems.
• Ability to own systems end-to-end in a fast-paced environment.

Responsibilities

• Own AI architecture and technical direction across the product.
• Build and ship production-grade LLM features with clear quality and performance metrics.
• Design and optimize AI pipelines across prompting, retrieval, agents, and evaluation.
• Lead development of agentic systems and multi-step tool use workflows.
• Establish standards for model evaluation, benchmarking, and observability.
• Partner closely with product, design, and engineering to ship high-impact AI features.

Skills

TypeScript

Python

LLM APIs

Prompt engineering

Agent-based AI

Software architecture

Tools

LoRA

Full fine-tuning

Distillation

Job description

We’re partnered with a well-funded early-stage startup rethinking how designers build in the AI era, backed by top-tier investors and operators from leading AI and technology companies. They’re building a next-generation design platform powered by LLMs, focused on helping product teams move from idea to production faster through intelligent, AI-driven workflows.

This is a foundational, high-ownership role where you’ll define and scale AI systems across the product. You’ll operate at the intersection of models, systems, and product, turning cutting-edge AI capabilities into reliable, measurable user-facing features.

What You’ll Do

• Own AI architecture and technical direction across the product
• Build and ship production-grade LLM features with clear quality and performance metrics
• Design and optimize AI pipelines across prompting, retrieval, agents, and evaluation
• Lead development of agentic systems and multi-step tool use workflows
• Establish standards for model evaluation, benchmarking, and observability
• Partner closely with product, design, and engineering to ship high-impact AI features

Requirements

• 5+ years of software engineering experience
• 2+ years working with LLM APIs and prompt engineering in production
• Strong proficiency in TypeScript or Node.js and Python
• Experience with LLM evaluation, benchmarking, and performance measurement
• Hands-on experience with fine-tuning (LoRA, full fine-tuning, distillation)
• Experience building agent-based AI systems
• Ability to own systems end-to-end in a fast-paced environment

What You’ll Build

• Custom smaller models tailored to specific design workflows
• Multi-step AI agent systems with tool use
• Real-time, streaming AI experiences
• Context-aware code generation and intelligent suggestions

If you're interested in building AI-native products that directly shape how modern teams design and build software, this is a rare chance to have early, meaningful impact on both the product and technical direction.