Akanthas
Put AI to work on waste management
For whom?
The solution for the waste management industry
For waste drop-off centers
To fulfill their mission, drop-off centers must cut unnecessary pickups, maintain high sorting quality and deliver transparent reporting — all at once.
With Akanthas, monitor each container’s fill level in real time to optimize rotations and bring down contamination rates.
With detailed, accurate statistics, track every site precisely and strengthen the traceability of your flows.
For sorting and treatment centers
Unloading inspections, waste characterization, dispute management and reporting tie up massive amounts of time, money — and risk.
With Akanthas, every drop-off is automatically inspected by our smart sensors.
Materials are identified, contaminants are flagged, and visual evidence is archived.
Need a deeper characterization on a specific flow? Our patented AI analyzes each item precisely and auto-generates compliant, actionable regulatory reports.
For engineering firms
In the field, manual waste characterization takes time, leaves room for error — and produces results that vary from one operator to the next.
With our Waste Quality solution, save precious time and raise the standard of your audits with:
- advanced AI-powered waste characterization in just a few minutes;
- automatic generation of precise, visual reports aligned with your clients’ regulations.
Stand out from the competition with characterizations that are faster, safer, and more reliable.
For construction sites and industrial facilities
On a construction site or industrial facility, keeping sorting and container pickups under control is critical to cut transport costs — and avoid penalty fees.
With real-time container monitoring, trigger every pickup at the right moment. No more overflowing containers. No more half-empty ones being hauled away.
Automatic sort-error detection helps you avoid reclassifications — and spot unjustified overbilling early.
Waste Monitoring
Cut waste management and transport costs with AI.
Akanthas sensors measure the fill level of your containers and detect sorting errors.
The result: pickups always triggered at the right moment — and reclassifications avoided.
- Automated pickup requests
- Fewer reclassifications
- €1 invested = €3 saved
Waste QualiCheck
- 80% fewer disputes
- 20 minutes saved per inspection
- 100% of incoming flows documented with visual evidence
Waste Quality
The solution that delivers advanced waste characterizations automatically — and generates the reports you need, on demand.
Our patented AI model recognizes more than 60 material categories and produces a compliance report that’s precise and ready to use.
Reduce risk for your teams and guarantee compliance across your operations.
- 100% compliant reports
- Characterization cost divided by 5
- +60 identifiable waste types
Raise your waste management game — starting today
Got a question about our solution?
Reach out or book a meeting with one of our advisors.
An all-in-one solution
Sensors
Software platform
They use Akanthas
- Waste Quality
- Sorting & treatment centers
What our customers say
Philippe Teulet
Nicolas Sansus
About
Who are we?
Since 2021, we’ve been building sensors and AI — driven by one conviction: waste only exists because we haven’t been able to see the true value of resources.
Our solutions turn waste management into a driver of environmental and operational performance. They support the transition to a truly circular economy.
Guided by our mission to give resources their value back, we move forward with rigor, responsibility and ambition.
Every day, we work to put technology at the service of concrete, measurable and lasting impact.
Download
our white paper
Today, 60% of professional waste ends up in landfill or incineration. In the face of environmental challenges, improving waste recovery is critical — for all of us.
Download our white paper to explore:
- The challenges facing the waste industry through 2030
- How regulations are evolving for professionals
- How AI can support companies
Author
Blog
Resources & news
Got a question?
Can your solution detect hazardous objects in waste?
Recognition capability depends on the type of object, how often it appears in your flows, and the quality of the images captured.
For common objects, our models are already trained. For more specific cases, we enrich the database using your feedback and images collected directly at your sites.
Today, our solution recognizes several categories of hazardous objects, such as:
- Aerosol cans
- Silicone cartridges
- Nitrous oxide canisters
- Car batteries, etc.
There are still some limits. Very small objects (for example a lithium battery) can be hard to identify, especially if partially hidden.
That said, our model keeps improving: the more real-world situations it sees, the more accurate it becomes. Your field feedback directly powers this continuous improvement.
If our teams are already doing a good job, what can your solution still bring?
Already having good results is a real strength — and it’s exactly what lets a solution like ours take things further.
We don’t replace people — we free up their time, strengthen data reliability, and reinforce traceability.
Even with strong teams, there’s always value in automating repetitive tasks, getting alerts when needed, and having objective evidence on hand in case of doubt or inspection.
That’s what makes the difference over time — especially as customer expectations, contract requirements and regulations keep ratcheting up.
How much does Akanthas cost?
The cost of an Akanthas solution depends on the volumes to monitor, your objectives, the type of flows, and your technical constraints.
We scope your needs to give you an accurate first estimate aligned with your priorities.
We offer you a fairly priced solution, genuinely adapted to your site — and capable of generating a solid return on investment.
How far can Akanthas' AI go in recognizing objects and materials in waste?
Our AI excels at the visual identification of standard materials and objects, with precision comparable to a human operator — but in real time and at scale.
Here’s what it distinguishes today:
- Material families: metals, plastics, glass, cardboard, wood, etc.
- Common objects: packaging, electronic waste, textiles, etc., based on shape and appearance.
- Visual anomalies: the presence of non-conforming materials in a flow (e.g., a tire in household waste).
What it doesn’t do (yet):
- Chemical or molecular analysis: it doesn’t distinguish between polymers (PP, HDPE, PS…) or complex compositions.
- Detection of traces or invisible contaminants.
- Fine recognition of rare materials: some materials (e.g., carbon fibers, composites) require specific training with field data.
Our AI improves with your real-world data: the more it’s exposed to your flows, the sharper it gets at picking up nuances (e.g., distinguishing a plastic film from a crumpled paper bag).
Got a question about our solutions?
