[2.7] Add Job Recipe API spec and design doc [skilp ci]#4195
Open
chesterxgchen wants to merge 5 commits intoNVIDIA:2.7from
Open
[2.7] Add Job Recipe API spec and design doc [skilp ci]#4195chesterxgchen wants to merge 5 commits intoNVIDIA:2.7from
chesterxgchen wants to merge 5 commits intoNVIDIA:2.7from
Conversation
Add docs/design/job_recipe_spec.md covering the Recipe API interface design: - User guide (constructor, export/execute, filters, config, decomposers, per-site config, model/checkpoint) - Execution environments (SimEnv, PocEnv, ProdEnv) - Memory management (server_memory_gc_rounds with per-recipe defaults) - 14 API enhancement requirements (deploy map, multi-script/launcher, k8s, algorithm config, optional integrations, resource/GPU, client tracking, job scope, client memory) - Proposed API changes for each requirement with recommendations
c590b5f to
49f2443
Compare
Contributor
Greptile OverviewGreptile SummaryAdded comprehensive Recipe API design document ( Major sections include:
The spec is a design document (not implementation) and correctly notes where proposed methods like Confidence Score: 5/5
Important Files Changed
Last reviewed commit: 5e919e0 |
…evel config - Section 10.7: Merge overlapping Options A+C into Option A (base Recipe method add_client_scripts); keep launcher type as separate Option B - Section 10.9: Distinguish job-level config (SwarmServerConfig/ClientConfig constructor args) from top-level config (add_server_config/add_client_config); add set_server_job_config/set_client_job_config proposal with conflict resolution policy (ValueError on overlap) - Section 10.11: Consolidate set_resource_spec/add_resource_spec into single Option B on base Recipe class
holgerroth
reviewed
Feb 13, 2026
- Strengthen migration note for deprecated BioNeMo decomposer pattern (recipe.job.to_server -> recipe.add_decomposers) - Add implementation note: add_client_scripts() is proposed; current impl has add_client_file() (single file) - Add implementation note: set_resource_spec() is a design proposal, not current API - Group Section 9 enhancements by target release (2.7.2 / 2.8.0) instead of flat numbered list
- Line 52: "(last section)" -> "(Section 9)" for API Enhancements ref - Line 328 was already corrected: "(Section 3.5)" -> "(Sections 3.5 and 3.6)"
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
docs/design/job_recipe_spec.md— the Recipe API interface design document covering:add_server_config/add_client_config), decomposers, per-site config, model/checkpointserver_memory_gc_roundswith per-recipe defaults from codebase)recipe.add_resource_spec(),recipe.add_client_scripts()on base Recipe class, and noting existingadd_server_config/add_client_configfor algorithm overrides)docs/design/README.mdindex fileTest plan
Made with Cursor