Overview Picnic's mission is to

drive a higher quality ad-funded internet . With 70% of people finding digital ads annoying, brands are wasting ad spend on ineffective and potentially damaging ad experiences. We're a fast-growing, founder-led start-up, passionate about making digital ads work better for everyone.

Now, we're looking for our first ever

Applied Machine Learning Engineer

to help us build intelligent systems that power the future of digital advertising.

Responsibilities

Build applied ML systems that make our product smarter and more defensible

Develop contextual categorisation of web content (e.g. automatically recognising industry categories)

Prototype algorithms that can improve advertising outcomes, such as smarter bidding or optimisation logic

Work closely with product and commercial teams to package outputs into demos and case studies for clients

Collaborate with engineers to integrate ML features into production system

Establish good practices for applied ML in the business, bringing your curiosity and drive to help us move fast while building robust foundations

Apply

modern LLM techniques

to business problems:

Use prompt-engineering strategies to get consistent, accurate results

Explore augmentation methods (e.g. combining LLMs with our own data via embeddings or retrieval)

Run fine-tuning experiments to adapt general models to our specific domain

Who are we looking for? We're looking for someone who is curious, hands-on and eager to shape the future of Machine Learning at Picnic. You'll be motivated by solving real world problems with cutting edge techniques and excited about working in a start-up where you'll have a big impact from

day one!

Ideally, you've got 1-3 years experience in applied ML/AI (or equivalent practical experience through academic projects or early career roles)

You're comfortable coding in Python and using ML frameworks (e.g. PyTorch, Hugging Face, scikit-learn)

You've worked with

large language models (LLMs)

- not just calling APIs, but experimenting with different ways to get the best out of them:

Prompt engineering

(designing effective prompts, chaining prompts)

Fine-tuning or instruction tuning

models for specific tasks

Embedding-based augmentation

(using vector search to give LLMs access to external knowledge)

Retrieval-Augmented Generation (RAG)

or similar techniques

You've taken messy, real-world data and turned it into useful, structured outputs

You can show how your work had impact - whether that's a model in production, a working prototype, or an experiment that unlocked a new direction

You're a clear communicator, able to work across functions with product managers, engineers, and commercial stakeholders

You're excited by the prospect of being our

first ML engineer , owning this area from day one and growing into a lead role quickly

You don't need to have worked in adtech before - but if you have, that's a bonus!

Benefits Why will you want to work for Picnic?

You'll be part of a culture that's grounded in our values and pushes us forward every day:

You're

trusted to take ownership , make decisions and deliver with autonomy

Ideas

move quickly

from concept to reality - we learn by doing, not over-polishing

We

constantly push ourselves and the industry forward,

aiming higher with every piece of work

Together, these values guide us as we tackle hard problems, learn fast and make digital ads better

What can Picnic offer you?

We're proud to have been recognised by Culture100 and Flexa as one of the best and most flexible small businesses to work for. We offer:

A flexible, hybrid working setup (we're usually in the office a few times a week, so being

within reach of London Bridge is important )

33 days holiday (inclusive of Bank Holidays), plus additional Christmas shutdown

Private Medical Insurance through Vitality

Picnic Pension Contribution

Inclusive Parental Leave Policy

A great co-working space, regular socials & offsites, Picnic Thursdays, Summer Fridays and Work from Roam opportunities

Salary Range:

£50-60k

#J-18808-Ljbffr