GeezerVison is showing devin.ai — Devin Desktop | Devin  New session Automations Security Review Wiki Customize No PR Widen the model hidden dimensionWorking... Closed Switch the MLP activation to GELU 42m ago· Display Status Space Pull request Agent Widen the model hidden dimension
Working... Learning rate tuning Increase the Muon matrix learning rate
 Working...Learning rate tuning Add a learning rate warmup phase
PR is ready• Learning rate tuning Tune the WSD warmdown ratio
Switch the MLP activation to GELU 42m ago• Establish the training baseline 3h ago• Add QK-norm and value embeddings  mainLaunchpad 0 0 Screen Reader Optimized Ln 231, Col 29 Spaces: 2 UTF-8 { } TypeScript JSX Cognition Platform (Enterprise) Windsurf - Settings One command center
for every agent and modelGet the best from every model and agent. Using the Agent Client Protocol (ACP), you can manage your favorite agents and all the frontier models in one place  implement backprop from scratch: a minimal scalar autograd engine, like micrograd I'll build a single `Value` type that records the computation graph as expressions run, then backpropagates with reverse-mode autodiff. Starting from an empty repo with the node scaffold: the data, its gradient, and the edges to its parents. Each forward op returns a new `Value` and stashes a local `_backward` closure, the chain rule for that op, pushing the output gradient into its inputs. Now the backward pass itself: topologically order the graph from the output, seed its gradient to 1, and run the stored closures in reverse so each node is fully accumulated before its parents are reached. Sanity-checking the gradients against PyTorch on a small expression with a reused variable, so the accumulation is actually exercised. python -m pytest test_engine.py -q Done. A ~50-line autograd engine, written from scratch. `backward()` topo-sorts the graph and applies each op’s local gradient in reverse, accumulating with `+=` so reused sub-expressions sum correctly. The gradient check matches PyTorch to 1e-6. A team of agents
for every engineerYou decide what to build, then agents write the code, chase the edge cases, and test every detail Migrate billing to Stripe v38 subagents · 16 PRs Your development environment,
everywhere you goClose your laptop, open another. Your environment is still running with Devin  1import { createServer } from "./http"; 2import { routes } from "./routes"; 3 4const app = createServer({ routes }); 5 6app.listen(8080, () => { 7 log.info("api ready on :8080"); 8}); Running in the cloud· API on :8080 Files5 running apps api server.ts billing-worker.ts web checkout.tsx theme.css db seed.sql docker-compose.yml .env.local package.json Migrate billing to Stripe v3 Migrate billing from Stripe v2 to v3 I split this into 4 subagents. Each one opens its own PRs. Rebuild billing settings pagePRs are ready  2Write checkout e2e testsPRs are ready 2Backfill invoice tax fieldsPRs are ready 2Coordinate your agent swarmComplete long running tasks with Devin coordinating subagents,
each producing its own PRs Fix flaky checkout test Fix the flaky checkout test, then run the full suite Fixed a race in the payment mock. The full suite runs on Python 3.10 to 3.13 and takes about 40 minutes. Running locally MacBook Devin Cloud Conversationsame context Branchfix-flaky-checkout Tests4 Python versions in parallel Handoff between local and cloud agentsRun sessions on your machine or in Devin Cloud, and move between them with the same context Manage any agent in one place Fast ContextWhere is reconciliation scheduled? Instant codebase contextFinds the exact files and lines your agent needs in milliseconds with Fast Context Built-in Devin ReviewRapidly review every diff using Devin Review from where you code    Fusion Pairs frontier intelligence with cost-efficient execution Cost-effective frontier resultsDevin Fusion pairs a frontier lead model with a cheaper sidekick that executes  117 def __init__(self, config): 120 self.window_sizes = self._compute_window_sizes(config) self.window_sizes = self._compute_window_sizes(config) 122 def _compute_window_sizes(self, config): 123 pattern = config.window_pattern.upper() 124 long_window = config.sequence_len 125 short_window = long_window // 2 126 char_to_window = {"L": (long_window, 0), "S": (short_window, 0)} 128 for i in range(config.n_layer): for i in range(config.n_layer): 129 char = pattern[i % len(pattern)] char = pattern[i % len(pattern)] 130 window_sizes.append(char_to_window[char]) window_sizes.append(char_to_window[char]) 131 return window_sizes return window_sizes Tab, Tab, ShipSupercomplete predicts your next thought, not just your next edit Teams building with Devin Desktop
Modal “Our engineers run multiple agents every day and Devin Desktop is the first tool that lets them manage all of them together, with shared context, from one place.”
Intact Financial “[Devin Desktop] is snappier, it's accessible, and it fits the way a lot of our developers are working today.”
NVIDIA “NVIDIA is joining Cognition's research preview for multi-agent support in Devin Desktop. Our engineers run multiple agents across complex workflows every day, and we're excited to help define how they share context and coordinate in one place.”
Harvey “With Devin Desktop's support for custom background agents, that context now extends to every engineer's laptop, so humans and agents work from the same shared understanding instead of starting from scratch.”
Ramp “Devin Desktop makes it easy to dispatch and monitor our array of agents from a single command center. We're excited to partner with Cognition to bring the agents Ramp engineers already use into one shared workspace, making it easier to jump between tasks, preserve context, and get more done.”
Make Devin Desktop your ownStop deprioritizingDo it all with DevinTrusted by developers.1M+UsersTrusted by over a million developers worldwide 4000+Enterprise CustomersStartups, agencies, and enterprises Frequently Asked QuestionsWhat is Devin Desktop? Devin Desktop is the new name for Windsurf. We’re building on the IDE foundation of Windsurf to introduce the command center for managing all your agents in one place. The Agent Command Center (Spaces, Kanban view, and multi-agent management) is front and center, while the full IDE experience you know remains fully accessible. How do I upgrade to Devin Desktop from Windsurf? Devin Desktop arrives as a standard over-the-air update, so your plan, pricing, extensions, and settings all carry over automatically. You can also download the latest Devin Desktop version from the download page. Will I lose anything if I update? The IDE, your extensions, workflows, settings, and in-progress work will all remain intact and will be fully migrated when you update. Only the name and branding are changing. Does my plan or pricing change? No. Your current plan and pricing stay exactly the same, including legacy Windsurf Enterprise plans. What is happening to JetBrains support? Windsurf for JetBrains (IntelliJ IDEA, PyCharm, WebStorm, and more) continues to be available for download. |
