Back to jobs

Principal Machine Learning Engineer

Pay
£150,000/year — pay source
Comprehensive compensation, including: Salary of £150,000 per year, with some flexibility based on experience. Annual salary reviews, ensuring they are competitive
Read the full posting
Work setting
Hybrid — work setting source
Time commitment: Full-time. Location: Hybrid is preferred, working 3 days in the London office (Monday, Wednesdays and Thursdays) and 2 days at home per week. We consider alternative requests if this is not possible for you. Reporting to: VP of Technology, Avishai Gurt
Read the full posting
Employment
Not stated

What you’ll work on

Full posting
  • Build a metric framework linking user experience and clinical efficiency to engineering metrics

  • Develop a clinician-defined quality standard for AI output using structured case review

  • Build evaluation infrastructure that measures AI output against the quality standard

  • Run experiments across the AI pipeline to attribute improvements and failures to specific stages

  • Make architectural decisions for the AI pipeline that serve the near-term roadmap

From the employer’s posting
A metric framework that connects Paloma's user experience and clinical efficiency goals to the technical metrics engineers optimise against day-to-day
A clinician-defined quality standard for AI output, developed with our Consultants through structured review of real cases
the evaluation infrastructure that measures against it
Experimental discipline across the AI pipeline, so that we know which changes caused which improvements and can attribute failures to specific stages
Architectural decisions on the AI pipeline that serve the near-term roadmap without foreclosing where we want to be in two years

What you’ll bring

All qualifications

Core experience

  • Deep experience of experimental design and error analysis: designing evaluations that isolate cause, attributing failures to specific components, and making architectural decisions on evidence
  • Experience in an operationally intensive business (logistics, marketplaces, on-demand services or similar) where metric definition was central to how decisions were made
  • Experience working directly with domain experts to convert tacit professional judgement into criteria a system can be evaluated against
  • Familiarity with systematic LLM evaluation techniques such as golden and regression datasets, LLM-as-judge pipelines, structured error analysis, human-in-the-loop review with domain experts
  • Deep understanding of designing and shipping agents and the harness around them: orchestration, tools and MCP integrations, context and memory management, and verification loops.
  • Experience working with health data and adhering to regulatory requirements
Qualification wording
Deep experience of experimental design and error analysis: designing evaluations that isolate cause, attributing failures to specific components, and making architectural decisions on evidence
Experience in an operationally intensive business (logistics, marketplaces, on-demand services or similar) where metric definition was central to how decisions were made
Experience working directly with domain experts to convert tacit professional judgement into criteria a system can be evaluated against
Familiarity with systematic LLM evaluation techniques such as golden and regression datasets, LLM-as-judge pipelines, structured error analysis, human-in-the-loop review with domain experts
Deep understanding of designing and shipping agents and the harness around them: orchestration, tools and MCP integrations, context and memory management, and verification loops.
Experience working with health data and adhering to regulatory requirements

About Paloma Health

Our mission is to make NHS autism and ADHD assessments and care accessible within 4 weeks of a GP referral, and, in turn, help every child achieve their potential.

In the employer’s words · Read in context

Apply at Paloma Health

Posting record

Report an error

Actively hiringFirst seen by us 6 days ago, still listed, and not a repost of an earlier posting.

Status in our records
Live
First seen by us
Nov 12, 20266 days ago
Recorded sightings
7
Last seen by us
Nov 19, 2026

These dates show when we found the posting. Check the employer’s website to confirm it is still accepting applications.

How this job compares

Time since we first saw it

Day 6

Early. Similar jobs usually come down between day 7 and day 75.

From 5,846 similar postings that came down. Coming down is not always a hire.

Paloma Health is hiring

2 open jobs

Open jobs by the week we first saw them, last 12 weeks.

Hiring record

See how this role fits your experience

Add your resume and we’ll mark which of these tools you already use, and point out anything in the posting that could rule you out.

  • Python
Where the posting names them
- Deep understanding of designing and shipping agents and the harness around them: orchestration, tools and MCP integrations, context and memory management, and verification loops. - Proficiency in Python, with a strong grasp of clean, typed, tested code and good software engineering practice - Excellent communication skills, including the ability to make a technical argument to a domain expert or commercial audience and to change your mind in public when the evidence says so

Your next step

Complete your application on job-boards.eu.greenhouse.io. The employer’s form will show what is required.

Applying through Greenhouse

  • No account needed: the form is one page, with an “Autofill my application” option that reads your resume.
  • Greenhouse doesn’t read resumes over 2.5 MB, or details placed in a header, footer, text box, table or second column. The file still attaches, but a recruiter has to type you in by hand.

Already applied? Track this application

Job description

View original posting ↗

Time commitment: Full-time.          

Location: Hybrid is preferred, working 3 days in the London office (Monday, Wednesdays and Thursdays) and 2 days at home per week.  We consider alternative requests if this is not possible for you.

Reporting to: VP of Technology, Avishai Gurt

About Paloma

Our mission is to make NHS autism and ADHD assessments and care accessible within 4 weeks of a GP referral, and, in turn, help every child achieve their potential.

In 2024, our founders, Mark and Darshak, were shocked to find that children needing an NHS autism or ADHD assessment were waiting 4-10 years for these assessments. Delayed assessments can result in children being more likely to develop mental health conditions and fall behind in education. Furthermore, the much-needed pre/post-assessment support (outside of ADHD medications) is typically not funded by the NHS. 

These long waits have built up because of increased recognition and understanding of neurodevelopmental conditions, and a lack of investment and innovation within the NHS to meet growing demand.

Mark and Darshak, with their 25 years of combined NHS experience, including building services that have treated over 1,000,000 patients in obesity & ophthalmology, founded Paloma to change this. 

As of May 2026, we have successfully reduced waits to less than 3 months by;

  • Taking a product-first approach to redesign the care pathway to give families a more consumer-like experience of care
  • Our in-house product, engineering and AI team building our own autism and ADHD specific electronic health record and AI clinical support tool to free clinicians from their significant documentation burden to refocus on supporting families
  • Investing in expanding the clinician workforce and upskilling them in new ways of working.

We are proud that Paloma is:

  • Loved by families, with 4.9 out of 5 Trustpilot and Google ratings
  • Care Quality Commission (CQC) registered and quality-focused
  • Clinically-led, with a team of over 170 people, including over 100 clinicians supporting families through their care journey. We are hiring 100+ more roles over the next 12 months.
  • A community of passionate, mission-driven and innovative team members who challenge the status quo and test and learn from new ideas and approaches.
  • In control of our own destiny, as we are profitable and backed by leading healthcare investors Triple Point and Heal Capital

What’s next?

  • Whilst we have reduced wait times, it is estimated that up to 500,000 children are still waiting for an NHS autism or ADHD assessment. We won’t rest until this comes down.
  • We want to build new services to support families and children diagnosed with autism or ADHD who feel lost after their assessments when the NHS is no longer there for them.

The Opportunity

Autism and ADHD assessment has no laboratory test and no external ground truth. What separates a good assessment from an adequate one lives in the judgement of experienced clinicians, applied case by case. Making that judgement explicit — specific enough to build against, and to prove that a system is meeting it — is the central technical problem at Paloma.

We are looking for a Principal Machine Learning Engineer to lead that work. This is the most senior individual contributor role in our AI squad. Your role is to define what good output looks like, how we measure it, and what a change has to demonstrate before it ships. 

What makes this role unique is that you will be designing this tool directly with our Consultant Child Psychiatrists who understand the potential complexity of each case. You can test each hypothesis with our data bank of thousands of completed assessments, then get immediate feedback from 100+ clinical team members on each live adjustment. Your subject matter experts are in our London office working alongside you, not on the other side of a contract.

You will work as a member of the AI squad, which sits within our wider product and engineering function alongside our expert clinicians, delivering:

  • A metric framework that connects Paloma's user experience and clinical efficiency goals to the technical metrics engineers optimise against day-to-day, and reporting against it to the Leadership Team
  • A clinician-defined quality standard for AI output, developed with our Consultants through structured review of real cases, and the evaluation infrastructure that measures against it
  • Experimental discipline across the AI pipeline, so that we know which changes caused which improvements and can attribute failures to specific stages
  • Architectural decisions on the AI pipeline that serve the near-term roadmap without foreclosing where we want to be in two years

As the AI team grows, this is the natural route into leading it.

Does this sound like you?

  • You treat measurement as the first discipline, not something added after the build. You are uncomfortable shipping a change you cannot demonstrate was an improvement.
  • You have defined metrics in a domain where none existed, and you are comfortable with the ambiguity that involves – including the judgement of when a rough measure now beats a perfect one later.
  • You can spend an hour with a domain expert and come away with something an engineer can build against, and explain to that same expert what the system can and cannot yet do. The translation matters in both directions.
  • You have raised the standard of how a team works, not only what it shipped. Colleagues adopt your habits because your reasoning is convincing.
  • You have strong judgment about what not to build, and you are willing to argue for it.
  • You take the responsibility of building for a clinical setting seriously and would rather get something right than ship it fast.
  • You have high agency, spotting what needs to be done and driving it through without waiting to be asked.
  • You embrace winning as a team, and enjoy positive, collaborative cultures

Responsibilities:

  • Own the metric framework for Paloma's AI systems: translate operational and product goals into technical metrics the squad optimises against, keep it current as the roadmap moves, and report progress against it to the Leadership Team
  • Define, with our Consultant Child Psychiatrists, what a complete and accurate AI output looks like for different case types, and what the threshold for evidence sufficiency is at each stage of the assessment pathway
  • Set the evaluation standard for the squad: specify what evaluation must exist for each part of the pipeline and what a change must demonstrate before it ships, working in close partnership with the AI Engineers who build and run those pipelines
  • Lead structured error analysis across the pipeline, attributing failures to specific stages rather than treating quality as a single aggregate number
  • Own machine learning architecture decisions, including model selection, agent and harness design and document the reasoning and evidence behind them
  • Establish metric-driven development as the default across the squad and the wider engineering team, through mentoring, technical review, and championing the practice with product and clinical colleagues
  • Document technical decisions and specifications in line with healthcare regulatory requirements

Qualifications and experience:

Required:

  • Substantial experience building machine learning systems within companies that reached real users, including at least one where you owned how quality was measured
  • A track record of defining metrics in an ambiguous domain — you have been the person who decided what to measure, not only the person who optimised against a target someone else set
  • Deep experience of experimental design and error analysis: designing evaluations that isolate cause, attributing failures to specific components, and making architectural decisions on evidence
  • Experience working directly with domain experts to convert tacit professional judgement into criteria a system can be evaluated against
  • Deep understanding of designing and shipping agents and the harness around them: orchestration, tools and MCP integrations, context and memory management, and verification loops.
  • Proficiency in Python, with a strong grasp of clean, typed, tested code and good software engineering practice
  • Excellent communication skills, including the ability to make a technical argument to a domain expert or commercial audience and to change your mind in public when the evidence says so

Desirable:

  • Experience in an operationally intensive business (logistics, marketplaces, on-demand services or similar) where metric definition was central to how decisions were made
  • Familiarity with systematic LLM evaluation techniques such as golden and regression datasets, LLM-as-judge pipelines, structured error analysis, human-in-the-loop review with domain experts
  • Experience working with health data and adhering to regulatory requirements
  • Experience with synthetic data generation or rubric-based evaluation for expert-judgement tasks
  • Experience mentoring AI engineers, or a period leading a small team

Why join Paloma?

For all employees, we offer:

  • A strong and positive culture that prioritises team building, transparency, and career development. This includes protected time and budget for team connection and learning.
  • Comprehensive compensation, including:
    • Salary of £150,000 per year, with some flexibility based on experience. 
    • Annual salary reviews, ensuring they are competitive
    • 5% employer pension contribution
    • Equity compensation through Share Options in Paloma Health 
    • Optional Private Medical Insurance and Employee Assistance Programme
    • Home Office Equipment Allowance
  • A unique annual leave benefits package which includes:
    • 37 days of annual leave. This is 28 standard days, plus eight bank holidays, plus your birthday off.
    • Option to buy an additional 5 days per year (giving a maximum of 42 days)
    • Where required, 2 days per year of special leave, for events that can happen in life, whether personal or family-related
  • Additional benefits include:
    • A premium performance company laptop
    • NHS discounts, including the Blue Light Card (as we are an NHS provider).

Equality & Diversity

Paloma Health is an equal opportunity employer committed to diversity and inclusion in the workplace and throughout the application process. 

Accommodations

We’re glad to accommodate candidates who need adjustments throughout the hiring process. We encourage you to be open about the adjustments you need, so we can adapt our interview process accordingly.

Safer Recruitment & Pre-Employment Checks

Keeping our staff and the families we serve safe is a top priority. As part of our pre-employment checks, you may be asked to complete a Disclosure and Barring Service (DBS) check. Depending on your role within the business, this will be at either an Enhanced, Standard, or Basic level.

Application process

Please apply via our website. We understand that applying for a new job requires significant effort, and we appreciate the time you put into it and look forward to reviewing your application.

 

Demand for ML Engineer roles

New ML Engineer postings at the employers we follow are up 12% over the last 6 weeks compared with the 6 before.

Weekly new ML Engineer postings at the same employers, from the week of August 17 to the week of November 9. The fewest in a week was 556, the most 1,080, and the latest week had 556.
August 17Last week

Each bar is one week of new postings. The darkest bars are the last 6 weeks.

Counted from the hiring pages we’ve followed for at least five weeks, so employers we add don’t look like growth. Updated weekly.

About the role

We file this job under ML Engineer, by the work the posting describes.

ML engineers put machine learning models into products. They write the code that trains, ships, serves and monitors models so a recommendation or prediction works for real users.

Postings ask for strong Python and software engineering, a framework such as PyTorch, cloud services, and more and more often experience with large language models.

A data scientist often works out which model to build; an ML engineer makes it run reliably in production.

US government figures for computer and information research scientists

Median yearly pay, May 2025
$140,300
Expected change, 2025 to 2035
21.8% more jobsagainst 3.5% for all jobs

The government has no code for ML engineers, so these figures are for the closest group, computer and information research scientists.

Source: US Bureau of Labor Statistics, Occupational Employment and Wage Statistics (May 2025) and Employment Projections (2025 to 2035). Occupation code 15-1221.

Source details

View original posting ↗

Source notes

Source excerpts

Selected passages from the saved posting. Check the full description for conditions and exceptions.

Pay
Comprehensive compensation, including: Salary of £150,000 per year, with some flexibility based on experience. Annual salary reviews, ensuring they are competitive
Location & working pattern

London

Time commitment: Full-time. Location: Hybrid is preferred, working 3 days in the London office (Monday, Wednesdays and Thursdays) and 2 days at home per week. We consider alternative requests if this is not possible for you. Reporting to: VP of Technology, Avishai Gurt
Work authorization

No clear work-authorization passage found. Eligibility is unconfirmed.

Find answers in the posting

AI
How answers work

AI selects complete passages from this posting. Check them for conditions and exceptions.

Uses this posting and your question. No profile needed.