rtkaleta
At Purple Dot, we are on a mission to create a world where, ultimately, every product that is manufactured is sold. The current e-commerce model leaves brands waiting for stock to arrive at the warehouse before they can start selling, leading to missed sales opportunities, unsold inventory and waste. We believe the next wave of innovation will happen in the time before stock arrives, enabling brands to sell inventory no matter where it is in the world. Our pre-commerce platform puts the power back in brands’ hands, allowing them to take control of their sales timelines and maximise revenue potential.
And we make pre-orders trust worthy for the shoppers too, keeping their money in escrow until items ship, and always up to date on where their pre-order is at, with an option to cancel no questions asked.
We've raised a $10m Series A in 2024, we've plenty of runway left and we're looking for help as we scale our platform through the next order of magnitude in terms of load on the system and ultimately the number of dollars we process.
We are a team of 22 split evenly between the UK and US. 32% female, including the CEO. All of us have a mix of startup and big tech experience. For the founding team this is our second (so far :) successful startup.
I have two engineering roles open:
* senior full stack engineer - based in London, UK - to join our core engineering team of 6 that are shipping our product roadmap: https://purpledot.notion.site/senior-full-stack-engineer-uk
* customer solutions engineer - based in New York, US - to work with our support and sales teams to help make going live with and using Purple Dot a delightful experience for brands and their shoppers: https://purpledot.notion.site/customer-solutions-engineer-us...
Like everything else we do, we interview fast, and you will deal with me (head of engineering) throughout.
The world of eCommerce was built to constraints of physical retail. So much so that e.g. in the US it is illegal to sell things online that aren't ready for dispatch from a local warehouse. Millions of brands building online stores on top of Shopify/Salesforce etc that have long manufacturing lead times have to predict demand then keep their fingers crossed that inventory will arrive on time weeks or months later ready for the marketing campaigns they had arranged.
This isn't working well - there is over $200 billion worth of dead inventory sitting in warehouses globally in fashion alone, and a staggering 25% of all eCommerce inventory is going unsold. To clear this brands hurt their gross margins by selling at discount or stuff just goes to waste. Radical change is needed to create a sustainable industry not just for our planet but for a thriving economy.
At Purple Dot, we believe the next decade’s most impactful innovations in solving this problem will take place before stock even reaches the warehouse. That’s why we power the operational infrastructure at scale, as well as fostering mission-critical consumer trust, to enable merchants to increase sell through by selling earlier. Whether stock is one day or six months away from the warehouse, our pre-order platform gives merchants the flexibility to always be selling and to never miss an opportunity to convert consumer demand.
Our solution is proudly trusted by hundreds of brands across nearly every industry including Mattel, French Connection, Lisa Says Gah and Knockaround processing millions of dollars per month in pre-order transactions.
We are a team of 6 engineers, 2 Sales folks, 2 Customer support folks + CEO/CTO located between London, UK and Charlotte, NC, USA – and are backed by investors like Connect, Unusual, and Moxxie Ventures. We are serial founders who have sold a start-up before (https://medium.com/@TwizooSocial/all-good-things-can-get-bet...). The use of our platform is growing triple digits on average YoY and so we are responsibly growing our engineering team to keep up.
If you fancy joining a fast growing startup, feel you meet the requirements and are located on the US east coast or in the London area, I'd love to chat. Our interview process is simple and we move fast. See job descriptions. I'm Ryszard (https://www.linkedin.com/in/rtkaleta/), I run our engineering team and the next step would be to speak to me f2f for 30 minutes to learn more about each other. Find me at [email protected].
In fact we had resolved to building a turker bot detector and started rejecting tasks completed in suspicious ways. Only once we have built ourselves a trusted turkers population did we start to get quality data back. I suspect most people don't bother to go back and reject poor quality answers and that is why the bots survive.
We use a multitude of models - linear and non-linear, supervised and unsupervised - to surface the best user-generated content from social for our clients.
Our ML stack is written almost entirely in Python - because of the readily available excellent libraries that give you many of the tools you'll need in your ML toolbox (matplotlib, numpy, pandas, scikit-image, scikit-learn, scipy to mention just a few). We have some more sophisticated image processing in C++. We find ourselves attacking quite a number of machine learning problems so to make our workflow more efficient we have built a layer of functionality above these libraries to manage the flow of data and speed up model prototyping and assessment. Processing input data and model fitting is usually done locally, with any hyper-parameter grid search or other computationally intensive tasks being run on a remote, cloud hosted cluster spun up on-demand.
Our tech is built around lambda architecture principles - we have a live path that processes new social content to the extent required to make content immediately available to relevant clients, and a batch path that processes complete datasets daily. Our ML models get used on both paths.
With a neat set of Python libraries for data manipulation and model definitions and our own model prototyping and assessment rig the biggest challenge is always building big, quality datasets. This continues to be our biggest learning and we are always keen to explore how others tackle this. We rely on a combination of in-house turking (which we consider a vital model development stage), external turking services (for volume) and have additionally looked at some knowledge transfer techniques.
See https://www.periscope.tv/w/1RDGlYXeDpOJL for a quick talk where I expanded on why and how Twizoo uses ML.