Glacier
"The future of recycling is Glacier."
San Francisco firm whose compact AI sorting robots retrofit onto MRF conveyor lines at up to 60 picks a minute, and sell SKU-level recovery data to brands.
Glacier, which operates on the domain endwaste.io, builds AI sorting robots and material-recovery data services for the recycling industry. The company was founded in April 2019 in San Francisco by Rebecca Hu-Thrams (CEO, previously a Bain & Company consultant) and Areeb Malik (CTO, previously a software engineer at Facebook), and employs roughly 90 people. Its stated goal is a world in which nothing valuable ends up in a landfill; in the United States only about a fifth of residential recyclables are actually recovered from a stream of roughly 70 million tons a year.
The core product is a compact pick-and-place sorting robot installed over the conveyor lines of material recovery facilities (MRFs). It occupies about the floor space of one worker and goes in without heavy machinery, structural retrofits or line shutdowns, which is the main commercial difference from the large optical-sorter and gantry systems it competes with. In production the robot performs up to 60 successful picks per minute at a reported 95 percent uptime across customer sites, and the company cites an average payback of about ten months. The vision models behind it are trained on some three billion images taken from real recycling operations and recognise more than 70 material categories, covering roughly 90 percent of a typical curbside stream: PET and HDPE containers, PE film, OCC, newsprint and mixed paper, aluminium and tin cans, foil, clear, green and amber glass, cartons, paper cups and individual items such as toothpaste tubes. Custom models can be trained for a specific item or branded SKU. A fiber-line variant adds clog resistance, and every unit ships with remote diagnostics, automated reporting and 24/7 operator support backed by field service across North America.
A second line of business, marketed as DataStream or Glacier Brand Intelligence, sells the item-level recovery data the robots generate back to consumer-goods companies. Coverage spans partner MRFs serving roughly one in ten Americans, and brands use it to substantiate recyclability claims under the FTC Green Guides, meet Extended Producer Responsibility rules, and support SEC climate and ESG disclosure. In a Colgate-Palmolive project the system verified 57,250 correctly sorted toothpaste tubes and a 198 percent year-on-year rise in tube recovery at one facility over twelve months. Reference deployments include RRRASOC in Michigan and Recology, a Detroit MRF recovering an additional 15 million PET bottles a year, and a California MRF that raised paper purity by 17 percent. Glacier also works with Amazon's Climate Pledge Fund on a pilot separating novel biomaterials from fossil-based plastics alongside Novamont, Natur-Tec and PepsiCo.
Funding began with a USD 4.5 million seed in April 2022 led by New Enterprise Associates, followed by a USD 7.7 million round led by Amazon's Climate Pledge Fund and a USD 16 million Series A announced on 29 April 2025 led by Ecosystem Integrity Fund, with NEA, AlleyCorp, Overture Climate VC, VSC Ventures, Elysium, Overlap Holdings, Cox Exponential, Alumni Ventures, One Small Planet and Working Capital Fund participating. Cumulative funding is reported at about USD 28 million across four rounds."
Primary type & automation activities this supplier delivers:
More related news
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
Contact Glacier
WEBSITE
https://endwaste.io/HEADQUARTERS
United States
Company Facts
Founded
2019
Primary Role
Manufacturer
Company Size
employees 50-100
Primary Region
-
Annual Sales
-
Funding Stage
Growth / Late-Stage VC
Funding Total
$ 28,200,000
Keywords
- Glacier
- Endwaste
- endwaste.io
- Glacier Robotics
- Waste management
- material recovery facility
- recyclables sorting
- contamination detection
- PET recovery
- fiber line purity
- SKU-level packaging data
- EPR compliance
- curbside stream
- Manufacturer
- Software/Algorithm
- Industrial Robot
- Perception & Vision
- Cloud & Data
- Manufacturing
