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
OpenAI

Applied AI Engineer - Startups

  • Office

Salary

Not stated

Similar roles pay £30K - 155K a year · our estimate

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 AI Deployment Engineering team works with frontier AI startups to make sure the company's technology is deployed safely and works well for them. The team also works with engineering, research, and product to turn what it learns into evaluation systems, product improvements, and better model behavior.

The team sits between what customers actually experience and how good the models are. It pairs hands-on technical work with product judgment, turning complex, high-value use cases into clear signals that can improve both the customer experience and the underlying systems.

You will help strategic startup customers push advanced AI further. You will work with some of the most closely watched AI startups, help them improve their own systems, and carry what you learn back into the company's research and products. You will dig into complex workflows, find the gaps that matter, and turn them into reproducible evaluations and technical insights that help shape the company's research and product direction.

The role suits engineers who are comfortable debugging a workflow, iterating on prompts or agents, designing evaluations, and working across research and product. You should enjoy ambiguous, high-impact problems and want to influence how advanced AI systems get better in practice.

Responsibilities

  • Work directly with strategic startup customers to understand their critical workflows, find failure modes, and spot high-impact opportunities for improvement.
  • Prototype and iterate on prompts, agents, and workflow designs to understand system behavior and create value for customers.
  • Turn real usage patterns into clear, reproducible evals, benchmarks, and technical artifacts for the Product and Research teams, so that customer learnings shape the roadmap and model development.
  • Build repeatable tools, patterns, and evaluation approaches that raise quality across multiple use cases.
  • Make sound judgment calls in ambiguous situations, balancing immediate problem-solving with longer-term system improvements.
  • Build relationships across the startup ecosystem and act as a technical partner to individual customers and the wider community.

Requirements

  • Strong software engineering and AI fundamentals. Past roles such as startup CTO, software engineer, ML engineer, data scientist, or equivalent are relevant. Experience shipping production systems end to end is a strong plus.
  • Experience as a technical founder or as an engineer at an early-stage startup.
  • Familiarity with, or interest in, model training pipelines and reinforcement learning.
  • Experience building AI applications, agents, or evaluation systems, and the ability to reason clearly about model behavior in complex workflows.
  • Comfort working directly with highly technical users and turning their challenges into concrete technical signals.
  • Ability to move easily between prototyping, debugging, evaluation design, and cross-functional work.
  • Clear communication with both technical and non-technical audiences.
  • High agency, strong product sense, and a preference for durable improvements over one-off fixes.

Conditions

The company is an AI research and deployment company focused on making general-purpose AI benefit all of humanity, and it deploys its systems through its products.

The company is an equal opportunity employer and does not discriminate on the basis of race, religion, color, national origin, sex, sexual orientation, age, veteran status, disability, genetic information, or other legally protected characteristics.

Background checks follow applicable law. For US-based candidates, qualified applicants with arrest or conviction records are considered in line with San Francisco, Los Angeles County, and California fair chance laws.

For unincorporated Los Angeles County roles, the company has found that criminal history may have a direct, adverse relationship with three duties: protecting entrusted computer hardware from theft, loss, or damage; returning all hardware and its data at the end of employment or assignment; and keeping proprietary and non-public information confidential. The role also requires access to secure systems and data, so an offer may be withdrawn on that basis.

Where you’d work

From the office

About the company

OpenAI

  • Industry: AI

Offices in London, United Kingdom, Berlin, Germany, Munich, Germany, Stockholm, Sweden, Paris, France

Also hiring in San Francisco, United States, New York City, United States, Sydney, Australia and 4 more places

4 of their 134 open roles are remote

Your chances

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

  • 19 checks run
  • 1 red flag

Still hiring?

13 checks

1 red flag

How crowded?

6 checks

No 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
Details9 facts · Role, Location, Compensation, Employment, Company
Type
Full-time
Industry
AI
Specialty
ML
Region
United Kingdom
Pay period
Annual
Show 4 more factsShow less

Role

Category
Data & Analytics
Specialty
ML

Location

Work model
Office
Region
United Kingdom
Offices
  • London, United Kingdom
  • Berlin, Germany
  • Munich, Germany
  • Stockholm, Sweden
  • Paris, France

Compensation

Salary
Salary by agreement
Pay period
Annual

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