· London · India · Pakistan · Series B · Now hiring

Building the operating system for the secondhand clothing economy

A $250 billion wholesale secondhand market still runs on WhatsApp. Fleek is the marketplace changing it, powered by AI.

Backed by

a16z
YC
Y Combinator
HV CapitalBurdaeBay

The world's most broken supply chain

When you donate clothes to a charity shop, most of them travel to Pakistan, India, or Africa to be sorted by hand, one item at a time. The ones worth selling get shipped back to stock vintage shops in the UK and Europe. The ones that aren't get burned or buried or sold for pennies.

That double journey exists because the technology to sort clothing locally has never existed. Fleek is changing that. We're the #1 vintage wholesale marketplace connecting resellers and retailers with secondhand suppliers around the world. Our technology grades, prices, and categorises clothing from a single photo.

Since 2022 we've helped save nine million items from landfill. We're a small, fast-moving team. If you want to work on something that actually matters, you're in the right place. [80% landfilled or incinerated (EPA)]

15B
items collected/year
70%+
shipped overseas
12%
returns to market
Bales of secondhand clothing in a sorting warehouse
9 million items
saved from landfill since 2022

Our impact since 2022

17,400t
CO₂ avoided
10.8B L
water saved
35M kWh
energy saved

The world's first LLM fine-tuned on secondhand fashion

FleekSort grades, prices, and catalogues inventory from a single photo, in 6 seconds. It's trained on the world's largest secondhand fashion dataset, and that dataset grows with every transaction on the platform. Trusted grades are what make the marketplace work: buyers across the US, Europe, Australia, and Japan can source at scale because they know exactly what they're getting before it ships.

FleekSort - AI Grading
AI GradingEstimated resale price: £12/item
Brand
Brand recognized
Authenticity score: 8/10
Predicted demand: 7/10
Style
Raw selvedge denim, heavy weight indigo
Bootcut silhouette, mid-rise, full length
Waist size 27, slim through the hip and thigh
~6slatency
82%accuracy
100tonnes processed / month

Automated Sorting

AI grading, sorting, and cataloging. Brand, quality, condition, desirability. From a photo, in seconds.

Agentic Sourcing

Buyers find exactly the right products instantly. Natural language search over AI-graded inventory.

AI Quality Check

Automated QC at scale. Every item priced and merchandised from real-time transaction data.

Why join

Second-hand is becoming the first choice. We're building the marketplace that makes it possible at scale

01

Build the infrastructure layer

A $250B wholesale industry has never had a proper marketplace. You're building it from the ground up: the technology, the supply, the market itself.

02

Work that matters

Fashion is the third highest polluting industry in the world. We're here to change that. We've saved 9 million items from landfill since 2022, and the numbers accelerate as you build.

03

Work on AI, with AI, ahead of it

You'll build real AI products and use AI to do your best work every day. We expect it in interviews. Your role will look different in 6 months. We want people energised by that, not unsettled by it.

04

Global from day one

Customers in the US, Australia, Japan, and across Europe. Suppliers in Pakistan, India, and Africa. The supply chain you're fixing is complex and global, and you'll touch all of it.

How we work

Non-hierarchical. High-ownership. Execution-focused

Dream Big and Disrupt Yourself

Push the most ambitious version of what you're building. Nothing is out of reach.

Absolute Ownership

Own your work end to end. Take it from problem to done, and take pride in the outcome.

Curiosity Leads the Way

Don't accept anything at face value. Ask questions you don't know the answer to.

Talk to the Customer

Every decision starts with the customer at the centre.

Embrace Diversity

A global team building for a global supply chain. London, India, Pakistan. Bring your authentic self.

Built by operators, not observers

Abhi and Sanket in a secondhand clothing warehouse
Abhi Arora

CEO & Co-founder

Early employee at Dubsmash (300M users, acquired by Reddit). UC Berkeley & Cambridge MBA.

Sanket Agarwal

CTO & Co-founder

Product & engineering at Google, Uber, Suki, and Postmates.

See the supply chain we're fixing

This is the world we're digitising. Every bale, every warehouse, every sorting floor. Follow along.

Working pattern

We're building something genuinely hard, and we do it best together. Our Shoreditch office is where the energy is: quick decisions, real collaboration, and a team that actually knows each other. One day at home for deep work. On average, people are in 3-4 days a week.

If you do your best work around people who move fast and care about what they're building, you'll feel at home here.

Our interview process

We move fast and keep things honest. Every stage is designed to give you real signal about who we are, how we work, and whether Fleek is the right next step for you. Here's exactly what to expect.

1
Intro call with our recruiter
A straight conversation about fit. We'll give you an honest picture of the role, where Fleek is right now, and what the next 12 months actually look like. Come with questions. This is as much for you to assess us as it is for us to learn about you.
2
Hiring manager conversation
You'll meet the person you'd be working with directly. They want to understand how you think, what you've built, and what you're proud of. This is a genuine conversation about your approach and ambition. Think about a problem you've solved that you'd want to talk through in depth.
3
Case study on site + meet the team
A practical exercise based on real challenges in the role, done at our Shoreditch office. You'll get a brief in advance so you can prepare. This is as much focused on collaborating and problem solving with the team, as it is seeing your case. After the case study, you spend time with the team -- lunch, informal conversations, genuine connection. Bring your questions about the culture, the roadmap, the things that matter to you.
4
Offer
If it's a yes on both sides, we move quickly. You'll always know where you stand and what the next step is before you leave each conversation.
Every candidate gets Claude Code

We give full access to Claude Code to every candidate who wants it, for the duration of our process. We know not everyone has it. We want you to come away from this process more AI-enabled than when you started.

Ready to build this with us?

All roles are based in London, Shoreditch. In-office 3–4 days a week.

Loading open roles...

Don't see your role?

We're always interested in exceptional people. Get in touch.

Get in touch