Understand the project
Navigate files, symbols and language-server context. Find the code that matters before making a change.
Built in Ethiopia. Available worldwide.
Build and ship scalable software.
With open-source models.
Turn ambitious ideas into precise work. Lamied brings AI-assisted development and native performance analysis together, so you can build with clarity and spot scaling bottlenecks before release.
Linux and Windows · Free editor. No account needed to download.
The Lamied workspaceFrom an idea to a running application
Code. Run. Review. Ship.Explore the workspace
The execution approach
Complex requests. Precise tasks.
Better instructions make smaller models more useful. Our architecture is designed to reduce ambiguity, unnecessary context and repeated work.
From request to execution
Follow the request as Lamied breaks it down, creates tools, and saves results.
One request enters the Mother Model.
Request and task referencesSaved results
Planned execution architecture. Output is saved on completion or interruption; task records persist until /eliminate.
Native performance analysis
See the cost inside your code.
A working feature can still hide a scaling problem. Lamied’s proprietary C++ engine estimates time and space complexity and detects repeated database and network calls.
Repeated scansO(u × o)
For every user, scan every order.
users.map(user => orders.filter(…))Indexed accumulationO(u + o)
Initialize users, then scan orders once.
totals.set(userId, current + amount)Illustrative comparison. A Map uses extra storage; expected complexity assumes average O(1) access. Static findings are estimates—confirm them with benchmarks.
14
Languages parsedNative syntax-tree analysis
5
Time & space analysisTypeScript, JavaScript, Python, C, C++
3
Cross-file call tracingTypeScript, JavaScript, Python
Built for the full development cycle
Edit, run and review your code.
Use the AI assistant to make changes across your project, run tests in the integrated terminal, and inspect the resulting diffs before you commit.
Build and run
Read and edit across the project. Run your application and focused checks in the integrated terminal.
Review and recover
Inspect file diffs, work with Git and return to checkpoints when a task needs a different approach.
Designed for more accessible AI development
Designed for lower AI coding costs.
Fragmentation gives smaller models focused tasks. Dynamic Tool Generation is designed to provide task references on demand and reduce repeated context. Our goal is a lower cost per successfully completed task.
Explore plansOur cost-reduction target50%
Lower AI cost per successfully completed task compared with current coding tools.
A development target. Not a measured result or a guaranteed saving.Model choice
Open models. Room to choose.
Use NVIDIA Nemotron through NVIDIA’s hosted API, choose another provider, or connect a local model through a compatible endpoint.
Model setup and requirementsBring your own model
Use your provider credentials or a local model. Available features depend on your Lamied plan and configuration.
One managed subscription Planned
Our launch plan brings model access into one Lamied subscription, with usage allowances and local-currency payments in Ethiopia.
Who is Lamied for?
Individual developers and small software teams building real products. Lamied combines a desktop development environment, AI assistance and native performance analysis. It is available worldwide, with Ethiopia as our first commercial market for the planned local-currency subscription.
What are Fragmentation and Dynamic Tool Generation?
They describe Lamied’s execution architecture: break a complex request into precise tasks, then give small models on-demand references to each task’s input and saved output. This is the direction described in our pitch deck; the task-tool warehouse is not yet part of the current download.
Is the 50% cost reduction proven?
No. It is our development target, not a measured saving or a guarantee. We plan to compare AI cost per successfully completed task while matching task quality, including token usage, retries and verification.
Which platforms can I use today?
Installers are available for Linux (Debian and Ubuntu on 64-bit Intel or AMD computers) and Windows (64-bit Windows 10 and 11). There's no macOS version yet.
Do I need an account or an API key?
You can download and use the editor without an account. AI features require an eligible Lamied plan. Depending on your plan and configuration, use Lamied-hosted models, your own provider API key, or a local model. Your provider bills any usage made with your own key.
Can I use a local model?
You can configure a local model through an OpenAI-compatible endpoint. Review embedding settings too: a local chat model alone does not make remote embeddings, sign-in, or other network-dependent features work offline.
Does my code stay on my machine?
Native parsing and local complexity analysis run on your machine. Cloud AI and remote embedding configurations may send code, prompts, or tool output to configured providers or services. Review those settings before using sensitive repositories.
Your next project starts here
Build with clarity. Ship with confidence.
Start with the free editor. Explore AI assistance and performance analysis on your own project.