You and the process

With a person
Anna, in-house recruiter
9 years hiring engineers
30 minutes with a real recruiter
They read your CV with you, on a call, and say where the offers are being lost.
Didn’t find what you were looking for? Tell us what to build
Preference Model

Member of Technical Staff - Research Engineer, Post-training

  • Office
  • 6+ years

Salary

$200,000 - 350,000/ year

AI summary

For members

The whole posting in a few lines. Sign up to read it here and on every role you open.

Sign Up to Read

Description

The company builds RL environments for training AI models that are more capable and better aligned, including future systems that may surpass human ability.

How a model behaves depends heavily on the reward signals used in training. Gaps between intended and actual behavior stem from limited capability or poor alignment, and both trace back to training objectives.

The team works on better RL environments for today's models and future ones, with the aim of pushing how far models can learn tasks they are weak at on their own.

The company has built RL environments for several frontier labs and has $16M in seed funding led by a16z. Its founders previously worked on Anthropic's data team, building data infrastructure and the datasets behind Claude.

This is a research and engineering role in post-training large language models, where you implement new approaches and help set research direction.

Responsibilities

  • Train and evaluate models on the company's proprietary RL environments to check data quality, find gaps in task coverage, and feed results back into environment design.
  • Build and run the RL training infrastructure, from training abstractions to distributed experiment management, using tools such as Verl, OpenRLHF, or similar, and help it scale to more complex research work.
  • Design, implement and test training environments, evaluations and methods for RL agents.
  • Profile training runs from data loading through reward computation, and speed them up to raise experiment throughput and shorten iteration time.

Requirements

  • You have run end-to-end LLM post-training pipelines on models of at least 7B parameters in industry.
  • You are proficient in Python and in PyTorch or JAX.
  • You have used at least one modern RL training framework.
  • You have built and operated ML infrastructure at scale.
  • You may also be a good fit if you have evaluated model outputs and built reward or evaluation signals.
  • You may also be a good fit if you keep up with post-training research and turn papers into working code.
  • You may also be a good fit if you hold firm but open views on structuring RL training code for reproducibility and fast iteration.
  • You may also be a good fit if you balance research exploration with engineering rigor.
  • You may also be a good fit if you have strong systems design and communication skills.
  • No PhD or large publication record is needed. Adaptability and strong communication and collaboration matter most for startup research.

Conditions

The role is onsite.

The company welcomes applications from people with diverse backgrounds and experiences, even if they do not meet every listed requirement.

Benefits

  • Cash and equity pay, positioned above the 90th percentile.
  • Ownership and autonomy in a fast-moving startup.
  • Working alongside top machine learning engineers.
  • Health, vision and dental coverage.
  • 401K match.
  • Lunch provided every day onsite.
  • Weekly snack orders.
  • Visa sponsorship and relocation support available.

Where you’d work

From the office

The office

Relocation offered

Visa sponsored

About the company

Preference Model

  • Industry: AI

Office in San Francisco, United States

Your chances

Worth a look before you spend an evening tailoring a CV for it.

  • 23 checks run
  • 4 red flags

Still hiring?

14 checks

1 red flag

How crowded?

9 checks

3 red flags

Fits Me

How well does this role fit you?

Answer a few questions or drop your CV, and every role gets a fit score with the reasons, this one first.

  • Your field
  • Level
  • Stack
  • Work model
  • Salary floor
  • Must-haves
Details15 facts · Role, Location, Compensation, Employment, Company
Tech stack
  • Python
  • PyTorch
  • LLMs
Seniority
Staff
Type
Full-time
Equity
Equity offered
Industry
AI
Specialty
ML
Show 9 more factsShow less

Role

Category
Data & Analytics
Specialty
ML
Seniority
Staff
Experience
6+ years
Tech stack
  • Python
  • PyTorch
  • LLMs

Location

Work model
Office
Region
United States
Office
  • San Francisco, United States
Relocation
Offered
Visa sponsorship
Sponsored

Compensation

Salary
$200,000 - 350,000 / year
Pay period
Annual
Equity
Equity offered

Employment

Type
Full-time

Company

Industry
AI

Something wrong with this vacancy?

Similar vacancies

Share this vacancy

What's wrong with it?

The employer never sees who reported.

Reason