Factory AI

Factory launched Agent Effectiveness, which measures agent usage against cycle time, work priorities and shipped artifacts so teams can see what their AI spend produces. The announcement gives no further detail; the demo video was not readable.

Factory added Connectors and Personas to Droid. Connectors are sign-in integrations with 50+ services, usable across Desktop, CLI, web and mobile, with MCP still available for private tools. Personas tailor starter tasks and recommended Connectors by role.

Factory announced that Grok 4.6 from SpaceX AI is now available in Droid, its coding agent. The post gives no pricing, rate limits or benchmark results, so the practical impact is model choice inside the tool.

Factory, an agent-native software development platform, has rolled out GPT-6 Astra to all customers, adding it alongside its existing model-agnostic model lineup for coding and dev workflows.

Factory's agentic coding tools join the Claude Marketplace alongside CrowdStrike, Cursor, and Vercel, letting enterprises apply existing Anthropic spend commitments toward buying Factory.

Factory compared single-agent vs multi-role systems rebuilding programs like gdal and 7-Zip from black-box behavior alone. Adding a role that defines a 'completion standard' before implementation raised gdal's behavioral parity from 36% to 90%, with similar jumps elsewhere.

Factory lays out a taxonomy of lint rules (grep-ability, glob-ability, architectural boundaries, security, testability, observability) meant specifically to give coding agents deterministic, machine-checkable feedback instead of vague natural-language conventions.

Factory announced Environments as generally available, positioning the sandbox an agent runs in as the foundation of its cloud agent experience. The linked details and video were not readable, so specifics beyond the announcement are unconfirmed.

Factory releases Droid 1.0, its model-independent agent harness, adding a programmatic tool-calling engine that lets models run hundreds of tool calls per request, plus an auto model router. Factory claims up to 20% better efficiency than native harnesses on the same tasks.

Factory's Droid desktop app adds in-app previews for documents, live websites, and full diffs, plus a design mode for commenting directly on the element you want changed.

Factory.ai built an evaluation framework to measure how well different context-compression strategies preserve useful information during long-running agent sessions (debugging, code review, feature implementation). Testing across real-world tasks, they found structured summarization retains more relevant context than alternative approaches from OpenAI and Anthropic.

Factory ships Connectors and Personas: Droid gets managed, sign-in-once access to the apps a team already runs — Linear, Slack, Sentry, PagerDuty, GitHub, Salesforce — with no server to configure, plus role-shaped onboarding. MCP still covers private and self-hosted systems in the same session.

Factory benchmarked thirteen models against the narrow job of reviewing pull requests, ranking them by cost against review quality rather than general coding ability — the tradeoff that actually decides which model sits in a review pipeline.

Factory makes Environments generally available. Sandboxes are defined once, prebuilt into images, rebuilt every few hours with validation, and reused by Automations and Missions under centrally managed enterprise controls.

Factory says OpenAI's GPT-6 Astra will reach a limited set of organizations, its own user base included, and will be selectable inside Droid through the Factory Desktop app or CLI. Timing is given only as soon.

Factory.ai's Missions feature lets AI agents work autonomously on large, multi-day software projects by breaking work into smaller units that fit within a single agent's context window. The architecture relies on separation of concerns and test-driven development applied at two levels to keep long-running, multi-step agent work reliable and verifiable.

Factory is acquiring HumanLayer, the coding-agent workflow team behind the Research/Plan/Implement approach. HumanLayer will be open sourced, its users invited to Factory, and the hosted platform kept running through a transition for export, migration or self-hosting.

Factory's Agent-Native Development framework argues coding agents' real bottleneck is now review, not writing code, and lays out concrete practices: write specs precise enough that success is demonstrable, keep tasks small, favor automated verification.

Factory replaced parts of its deterministic secret scanner with two fine-tuned models, one tuned for recall on secrets the pattern-matcher misses and one for clearing its false alarms, reporting frontier-comparable accuracy at a fraction of the cost and latency.

Factory launched a native desktop app for running multiple Droids at once, each with a persistent cloud or local computer, direct control over VS Code and other desktop apps, and support for MCP, skills, hooks and plugins across sessions.

Model prices among agent-capable models span fiftyfold, and one task can burn hundreds of calls — so where the routing decision lives matters. Factory's argument: the gateway sees a completion request, the harness sees a computation that runs for hours, and only it can weigh a model switch against t

Factory ships redesigned Analytics for its agent platform, reporting consumption, model mix, cost concentration and adoption by user, with an Analytics API and a claimed 63% cost reduction from its model router.

Factory's Deferred Context Engine loads MCP tool schemas, skills, and plugin instructions only when a task actually needs them instead of upfront, cutting measured input tokens by 15% on average and up to 50% in sessions with over 100 hidden tools.