経路Routing
Send every request to the model that gets it right at the lowest cost — with per-tenant policy and drift detection.
- 14 models, one policy
- Per-tenant cost caps
- Drift-aware reroute
序A research lab that shipsThe space between
The space between
thinking and acting.
Blossom AI is a research lab and a product company. We study how models should reason, route, and collaborate with experts — and we put that work into systems that operate reliably inside operationally complex businesses.
思考と行動の間·The space between thinking and acting
製品の角度How we build products
Four angles. One platform.
01
経路Routing as a surface
Every request goes to the right model — with per-tenant policy and drift-aware reroute.
02
校正Calibrated under load
We grade refusal, deferral, and uncertainty as first-class outcomes — not failures swept under the rug.
03
評価Governed evaluation
Curated datasets and LLM-as-a-Judge under governance you can audit.
04
監査Auditable end-to-end
Every decision is policy-versioned, replayable, and inspectable by the team that has to defend it.
二層Two layers, one disciplineA lab and a product.
A lab and a product.
Each makes the other honest.
Research without deployment becomes performance art. Product without research becomes a wrapper. We hold the two together — what we learn in the lab ships into our products; what we see in production sets the lab’s next question.
研究室LabsResearch
How should AI systems learn, reason, and collaborate with human experts?
Blossom Labs studies the open questions underneath modern AI deployment — scalable knowledge discovery, calibrated reasoning, and reinforcement learning in simulated operations. We publish what we find.
- 推論Reasoning under distribution shiftHow models stay calibrated when the world moves.
- 協働Human–expert collaborationWhere to ask, where to defer, where to act.
- 模擬RL in simulated operationsPractice in environments before production.
製品ProductDeployments
Three systems that put research-grade AI into operationally complex businesses.
Logistics, finance, manufacturing, healthcare ops — domains where a small share of bad decisions costs real money or worse. Operators get routing, evals, and agents that hold up under load.
- 経路RoutingEach request to the right model — cost, latency, policy.
- 評価EvaluationProduction evals on real traffic; regression attribution.
- 行動AgentsLong-horizon work with human approvals where they matter.
製品The product
All product docs Three systems that work together.
評価Evaluation
Production evals on real traffic distributions. Catch regressions before they reach users; attribute them to the change that caused them.
- Live traffic replay
- Attribution-first
- Regression budgets
行動Agents
Long-horizon agents for operationally complex work. Human approvals where they matter, autonomy where they don't — with auditable traces.
- Approval policies
- Tool sandboxing
- Replayable traces
流れHow it flowsFrom research question
From research question
to live operations.
We don’t separate “ideation” from “delivery.” Every loop closes back into the lab.
01問い
Question
A research question rooted in something we saw in real production.
02実験
Experiment
Simulations, evaluations, ablations. We publish what we find.
03橋渡し
Crossing
Findings become operating principles — codified, peer-reviewed, versioned.
04実装
Implementation
Principles ship into Routing, Eval, or Agents — gated by an internal eval bar.
05運用
Operations
Production telemetry feeds back; the lab's next question begins.
研究Recent work
All research From the lab.
展望 · Vision
The Last Mile: How Blossom Lab Bridges GenAI and SME Businesses
As we enter 2026, the AI landscape has transformed dramatically. But for SMEs, the gap between cutting-edge AI and practical business value remains. Here's how Blossom Lab is building the 'Last Mile' connection.
事例 · Case Study
Case Study: Fully Booked - Solving Impossible Bookings
How we built an intelligent concierge platform that unlocks fully-booked luxury hotels in Japan.
技術 · Technology
Why Reinforcement Learning Matters for Business
Understanding the competitive advantage of RL in modern enterprise decision-making and optimization.
研究室より
The questions worth answering only show up under load — and the products worth shipping demand the rigor of a lab.
From the lab — Apr 2026
研究室の現況
What the lab is working on now.
- Routing policies under multi-tenant drift
- RL agents in simulated logistics environments
- Expert-in-loop training for Operations
- Calibrated refusal for long-horizon work