Answering "can this site take 120 MW" means reading a tariff PDF, an ISO queue spreadsheet, a transmission plan, and a state docket that disagree on what the utility is even called. Gridstate normalizes them into an API where every field cites its source document and page. Facts are bitemporal, so you can ask what a tariff said on a date and separately what we knew on a date. Capacity comes back as utility confirmed, published, indicative, or unknown. Most sites are unknown.
AI coding agent spend is scattered across every developer's machine: each agent keeps its own logs, in its own format, under its own plan. Tokenbase reads those logs where they already live and rolls them up into spend and usage by team, with Claude Code, Codex, Copilot, and Cursor normalized to one basis so the numbers actually compare. Everything is measured from the agents' own logs, not modeled from estimates. You can even use it to help you get hired!
Every TypeScript agent SDK I tried treated Azure OpenAI as a second-class citizen. Stratus flips that. It's built Azure-first and includes everything you'd expect: tools, subagents, handoffs, sessions, streaming, structured output with Zod, guardrails, hooks, tracing, and MCP client support. Plus Code Mode, which lets the model write code to orchestrate tools in a loop instead of one call per turn. If you're on Azure and tired of adapter tax, this is for you.
