Customer Success Stories

How Kiwibot Automates Data Curation for Autonomous Delivery Robots with Lightly

Lightly helped Kiwibot boost segmentation accuracy by selecting only the most valuable training images from millions of robot-collected frames.

Carlos Alvarez
Machine Learning Engineer
Overview

Lightly helped Kiwibot boost segmentation accuracy by selecting only the most valuable training images from millions of robot-collected frames.

Industry
Robotics
Location
Berkley, USA
Employee
>100

Get Started with Lightly

Talk to Lightly’s computer vision team about your use case.
Book a Demo
Products
Lightly One
Results
3x
Faster Model Iteration Cycle
Use Case
Data curation for autonomous robots

About

Kiwibot’s autonomous delivery fleet operates across complex pedestrian environments — dynamic settings where sidewalks, roads, and seasonal conditions are constantly changing. Each robot continuously captures images from multiple cameras, uploading them to the cloud — generating millions of images for potential use in training their segmentation models.

Problem

However, collecting data wasn’t the hard part — the real challenge was deciding which images were actually worth labeling.

Their models needed to adapt quickly, but labeling is expensive, and selecting the right data to annotate was tedious, manual, and error-prone. 

Before Lightly, Kiwibot relied on GPS data and random sampling to select images for training. The process was time-consuming, repetitive, and often surfaced redundant or uninformative scenes - especially in environments like long, uniform sidewalks or intersections with no clear boundaries.

Testimonials

"We had millions of images but no clear way to prioritize. Manual selection was slow and full of guesswork. With Lightly, we just feed in the data and get back what’s actually worth labeling."

Carlos Alvarez

Machine Learning Engineer

Scalable and Efficient Data Curation using Lightly

Kiwibot replaced this manual selection process with a programmatic pipeline built around Lightly Worker, Lightly’s code-first data selection engine. 

The team integrated Lightly directly into their data pipeline, enabling automated selection of the most diverse and informative samples from millions of robot-captured images. Their system now:

  • Combines model uncertainty signals, camera metadata, and Lightly’s diversity-based sampling to flag edge-case images
  • Runs continuously, fully integrated into the training workflow — no human-in-the-loop selection required
  • Outputs only high-value frames for annotation, reducing label waste and improving iteration speed

Lightly’s embedding visualizations also gave early confidence that selection was working well, while Kiwibot appreciated the hands-on support and constant updates from the Lightly team.

“Lightly isn’t just a tool - the team actively shares updates, model suggestions, and helps us move faster. It’s not something you get from most companies.” - Carlos Alvarez

Results

Kiwibot selected Lightly because it directly addressed a critical problem: the inability to consistently identify and label edge cases that matter. With over a million images coming in from their robot fleet, randomly sampling data wasn’t cutting it — it often resulted in redundant, uninformative samples that did little to improve model performance.

Lightly enabled them to shift from reactive to targeted retraining by automatically surfacing the types of data that cause real-world failures in production.

Common edge cases include:

  • Sidewalks merging into roads without clear visual separation, confusing the segmentation model
  • Snow-covered surfaces, which dramatically change visual features and scene structure
  • Puddles and potholes, often misclassified due to lack of representation in the training set

These scenarios were nearly impossible to catch with GPS-based or heuristic sampling. With Lightly’s selection pipeline plugged into their training loop, the team can now systematically surface these failures — leading to faster, more focused retraining cycles.

Get Started with Lightly

Talk to Lightly’s computer vision team about your use case.
Book a Demo
Testimonials

What engineers say after adopting Lightly

No fluff—just results from teams using Lightly to move faster with better data and models.

"Through this collaboration, SDSC and Lightly have combined their expertise to revolutionize the process of frame selection in surgical videos, making it more efficient and accurate than ever before to find the best subset of frames for labeling and model training."

Margaux Masson-Forsythe
Director of Machine Learning

“Lightly enabled us to improve our ML data pipeline in all regards: Selection, Efficiency, and Functionality. This allowed us to cut customer onboarding time by 50% while achieving better model performance.”

Harishma Dayanidhi
Co-Founder/ VP of Engineering

“By integrating Lightly into our existing workflow, we achieved a 90% reduction in dataset size and doubled the efficiency of our deployment process. The tool’s seamless implementation significantly enhanced our data pipeline.”

Usman Khan
Sr. Data Scientist

“Lightly gave us transparency to a part of the ML development that is a black box, data. Furthermore, Lightly enabled us to do Active Learning at scale and helped us improve recall and F1-score of our object detector by 32% and 10% compared to our previous data selection method. We finally saw the light in our data using Lightly.”

Gonzalo Urquieta
Project Leader

"Lightly is hyper-focused on finding thousands of relevant images from millions of video frames to improve deep learning models. The Lightly platform enabled us to build models and deploy features more than 2x faster and unlock completely new development workflows."

Isura Ranatunga
Co-Founder and CTO

"I was truly amazed once we received the results of Lightly. We knew we had a lot of similar images due to our video feed but the results showed us how we can work more efficiently by selecting the right data"

Alejandro Garcia
CEO

Explore Lightly Products

Lightly One

Data Selection & Data Viewer

Get data insights and find the perfect selection strategy

Learn More

Lightly Train

Self-Supervised Pretraining

Leverage self-supervised learning to pretrain models

Learn More

Lightly Edge

Smart Data Capturing on Device

Find only the most valuable data directly on device

Learn More

Ready to Get Started?

Experience the power of automated data curation with Lightly

Book a Demo