Detonator
Detonator: detecting backdoored clients in federated LLM instruction tuning by amplifying their own updates
Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.
Want to reach these readers? Sponsor an issue.
GITHUB · launched Nov 17, 2026 · 11 upvotes (top 98%) · 0 views
Niche: llm security and guardrails software · 65 launches, 7 in the last 7 days · Explore this niche →
Category and tags
- Horizontal
- Observability & eval
- Developer
- AI-native
- Hobby / open-source project
Visit website ↗ Source listing ↗
More like this: all 65 launches in llm security and guardrails software →
Could you build this?
No Detecting backdoored clients in federated LLM instruction tuning is novel security research, not a product that can be assembled from known components.
What it would actually take: A real version requires a federated learning testbed with LLM fine-tuning across many clients, plus the detection method itself, which must be designed and evaluated against adaptive attacks. This needs ML security research expertise and substantial GPU compute.
Competitors
Other products that read as similar to this one : 4057 launches clear the similarity bar, closest 8 shown.
Attention rank: #3990 of 4058 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).
Launched 384 days after the earliest competitor.
- Sentry · github · 2026-10-02 · 65 upvotes · similarity 0.63
- ReasonGate- An explainable gate that blocks LLM prompt injection · hn · 2026-07-16 · 7 upvotes · similarity 0.60
- Sentrint, a security scanner for projects build with LLMs · hn · 2026-08-20 · 6 upvotes · similarity 0.57
- memory-bench-registry · github · 2026-10-11 · 153 upvotes · similarity 0.56
- REx-skill · github · 2026-09-23 · 96 upvotes · similarity 0.55
- borg · github · 2026-11-07 · 15 upvotes · similarity 0.55
- Cloclo · hn · 2026-04-06 · 5 upvotes · similarity 0.55
- Axe · hn · 2026-03-03 · 6 upvotes · similarity 0.55
The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.