Programming languages
Explore language design that makes software clear, safe and fast to build. Blueberry's .bb direction is an early environment for experimenting with syntax, types, tooling and runtime ideas.
.bblanguage designtype systems
Research program · 2026 → forever
Explore
what's next.*
Blueberry Research explores technologies before they become products—starting with programming languages, AI and computing systems, then expanding toward the physical world.
Research should create options for the future.
Current focus
Build foundations first.
The earliest Blueberry research is deliberately concentrated. Strong languages, tools, AI systems and infrastructure can become reusable foundations for nearly everything that follows.
Compilers & developer tools
Study parsers, intermediate representations, diagnostics, build systems, package workflows, language servers and the full path from source code to a useful developer experience.
compilertoolingDX
Artificial intelligence
Investigate small and large models, local inference, training, memory, tool use, multimodal interaction, evaluation and AI systems that can meaningfully assist people.
Human-computer interaction
Explore interfaces that reduce friction: adaptive systems, better creative tools, natural interaction, accessible design and software that feels simpler without becoming less capable.
Cloud & distributed systems
Research identity, synchronization, storage, edge computing, reliability and service architectures that can connect Blueberry products without unnecessary complexity.
Research map
From code to the physical world.
Not every field below is an active product program. The map describes where Blueberry Research can expand as the company gains the expertise, capital and technical foundations to explore responsibly.
Languages
Syntax, semantics, types, runtimes and new ways to express software.
AI
Models, agents, memory, inference, training, evaluation and human-AI collaboration.
Operating systems
Computing environments designed around modern development, intelligence and devices.
Distributed computing
Edge systems, resilient services, synchronization and globally available software.
Hardware
Devices, sensing, embedded systems, new form factors and human interfaces.
Silicon
Accelerators, controllers, SoCs, hardware/software co-design and efficient compute.
Robotics
Simulation, manipulation, mobility, autonomy and useful machines in the real world.
Communications
Networks, connectivity, resilient communications and new infrastructure models.
Health technology
Wellness systems, sensing and—only with mature safety and compliance—regulated health technologies.
Energy & materials
Materials, manufacturing, energy systems and technologies that make physical products better.
Transportation
Intelligent mobility built on software, AI, compute, sensing, robotics and energy.
Aerospace
Aviation, autonomous systems and advanced research where Blueberry's earlier technologies may eventually converge.
Long horizon
Research ahead of products.
These years are directional gates, not promises or deadlines. Later work should move only when earlier stages have earned the resources and expertise required.
Programming languagesLanguage specification, parser, runtime, diagnostics and foundational tools.
Compilers & toolingCompiler architecture, IDE foundations, package systems, debugging and developer workflows.
Artificial intelligenceAI products, local and cloud inference, model evaluation and practical intelligent systems.
Operating & distributed systemsCloud foundations, synchronization, deployment, APIs and experimental operating environments.
Hardware & siliconDevices, embedded systems, FPGA work, accelerators and eventually custom silicon research.
Robotics & communicationsSimulation, autonomous machines, networking and the infrastructure connecting intelligent devices.
Health, energy & materialsSafety-first health exploration, sensing, manufacturing, energy and new material systems.
TransportationResearch into mobility systems where AI, robotics, compute, communications and energy converge.
Aerospace & advanced researchLong-horizon work driven by genuine technical advantage—not expansion for its own sake.
Principles
Useful curiosity.
Research should be ambitious without becoming detached from reality. These principles keep Blueberry's exploration grounded.
Build to learn.
Prototypes produce sharper questions than speculation alone. Make the smallest artifact that can test the idea.
Measure what matters.
Define the claim, baseline and evaluation before declaring an experiment successful.
Stay legible.
Research is more valuable when others can understand the assumptions, methods, limitations and results.
Safety scales with capability.
As systems become more powerful or enter regulated physical domains, safety, security and governance must mature with them.
Reuse foundations.
Prefer technologies that create leverage across multiple future products rather than isolated demonstrations.
Earn the next horizon.
Move into harder fields only when Blueberry has the resources, expertise and reason to do so responsibly.
Method
Question → prototype → evidence.
A lightweight research loop keeps projects moving without confusing experiments with production commitments.
Question
State a precise problem or hypothesis and why it matters.
Prototype
Build the smallest credible implementation that can reveal something new.
Evaluate
Compare against a baseline with tests, measurements, failure cases and limitations.
Transfer
Publish, archive, open-source or transfer useful findings into a product team when ready.
“Blueberry Research is where we explore technologies 5–10 years before they may become products.”