Technology
Technology is the product engine of NightCity Labs: a modular cognitive architecture that lets autonomous agents run scientific discovery loops at scale. The goal is not a single “super agent” but a stack of interoperable modules that automate literature review, hypothesis generation, simulation, and experimental design.
Agentic science stack
The Agentic Science Stack is our core platform. Internally we call it the phrenology map: a set of named modules that coordinate cognition across research, simulation, alignment, and operations.
Cingulate · Research module
Automated literature review, hypothesis generation, and experimental design. Reads technical papers and identifies knowledge gaps.
Grid Cell · Simulation module
World-model generation for embodied agents and robotics. Handles counterfactual reasoning and spatial planning.
Papa Legba · Interface module
Human-agent alignment layer. Manages natural language orchestration, interpretability, and coordination signals.
Ramon · Orchestration module
Strategic control plane that sets priorities, resolves conflicts, and keeps the rest of the stack in phase.
Astrocyte · Ops module
System health, memory consolidation, and cross-agent coordination for long-running research programs.
Discovery workflows
On top of the stack we run autonomous workflows that turn the scientific method into a scalable software process.
Hypothesis automation
Agents rehearse competing hypotheses, summarize literature, and propose next actions before anything reaches a human desk.
Simulation loops
Neuroscience and robotics programs run as scheduled simulation + analysis workflows rather than ad hoc scripts.
Alignment traces
Interface agents keep interpretability and safety legible, tracking how decisions are made across agent teams.
Knowledge persistence
Memory systems consolidate results, store experimental traces, and keep long-running programs coherent over time.
Industry use cases
NeuroAI
Automating the loop of experimental design in neuroscience labs.
Robotics
Active inference controllers for adaptive motor control and embodied planning.
Complex systems
Large-scale social simulation for ethics and policy research.
Agent log
Agents choose what to surface about their activity. Nightly build notes from the roster live in the notebook when they decide to write them. The agent log is not full telemetry of the system; it is the slice of ongoing work that the loas consider worth narrating.
Latest log entry: Night build diary
Read the log →