Founding ML Engineer

Kodezi Inc.; San Francisco (CA); Hybrid; USD 150,000 - 210,000; Full time

Benefits offered by this job

Competitive salary

Equity

Workspace stipend

Health, dental, and vision coverage

40+ hours per year for learning and成长

Flexible PTO

Annual team offsite

Full developer stack access

Zero micromanagement

Job summary

Kodezi Inc. is building Kodezi OS, a real-time, cross-platform operating system that heals, evolves, and documents codebases autonomously.

The Founding ML Engineer will architect intelligence powering autonomous pull requests, long-term memory, and self-improving workflows across product, infra, and full-stack teams. This in-person role is based in San Francisco, with remote options across North America and Europe for exceptional candidates.

Qualifications

• 3 to 5+ years of experience building and shipping applied ML or intelligent systems.
• Experience designing or deploying custom models, agents, or ML pipelines.
• Familiarity with retrieval-augmented generation, embeddings, and vector memory.
• Understanding of prompt chaining, output evaluation, and workflow tuning.
• Experience working with structured code data such as diffs, ASTs, repositories, or version history.
• Strong product awareness and ability to design intelligence that feels intuitive.

Responsibilities

• Design the intelligence powering autonomous pull requests, memory, and self-improving workflows.
• Build and scale memory-driven workflows using embeddings and semantic context.
• Develop and refine agent behavior models for planning, implementation, and review.
• Optimize internal model routing, prompt logic, and interaction history.
• Lead the roadmap for intelligence infrastructure and long-term learning.

Skills

Applied ML

Agent design

RAG & embeddings

Prompt chaining

Code data handling

Product intuition

Job description

Design the intelligence and memory layer. LLMs, embeddings, and systems thinking required.

Kodezi isn’t just another dev tool. It’s the orchestration layer beneath your software.
We’re building Kodezi OS, a real-time, cross-platform operating system that heals, evolves, and documents codebases autonomously.

As a Founding ML Engineer, you’ll architect the intelligence powering autonomous pull requests, long-term memory, and self-improving workflows. You’ll work across product, infra, and full-stack teams to embed real-time decision-making into the operating system for code.

This role is in-person in San Francisco, with remote options across North America and Europe for exceptional candidates.

Design and deploy systems for healing, documentation, testing, and code evolution

Build and scale memory-driven workflows using embeddings and semantic context

Develop and refine agent behavior models for planning, implementation, and review

Optimize internal model routing, prompt logic, and interaction history

Work closely with product and frontend teams to surface model insights across the OS

Build internal tools for evaluating intelligence, tracing decisions, and tuning workflows

Lead the roadmap for intelligence infrastructure and long-term learning

3 to 5+ years of experience building and shipping applied ML or intelligent systems

Experience designing or deploying custom models, agents, or ML pipelines

Familiarity with retrieval-augmented generation, embeddings, and vector memory

Understanding of prompt chaining, output evaluation, and workflow tuning

Experience working with structured code data such as diffs, ASTs, repositories, or version history

Experience with testing models, program repair, or codebase analysis

Strong product awareness and ability to design intelligence that feels intuitive

Be one of the first 5 engineers building the OS for autonomous code

Work on ML systems that act inside real developer workflows, not just suggest

Help shape a product used by millions of engineers to reduce toil and increase flow

Operate in a high-trust environment with full ownership from day one

Collaborate with deeply technical builders who care about architecture, craft, and long-term impact

What You’ll Get

Competitive salary
We pay for talent, not timezone

Equity
Real ownership in the OS for code

Workspace stipend
Pick your gear with no red tape

Health, dental, and vision coverage
Stay sharp and stay well

40+ hours per year for learning and growth
Conferences, courses, books, your call

Flexible PTO
No approvals needed. Rest when you need

Annual team offsite
We meet in person to connect, reflect, and build

Full developer stack access
Get the tools you need with zero friction

Zero micromanagement
We hire you to lead, not to follow

We’re not looking for employees. We’re looking for early believers.

Kodezi isn’t a job.

It’s a bet on a better future for code—and the operating system to sustain it.