feat(devx): local dev setup for control plane and full end-to-end flow (MLI-6681)

[lilyz-ai]

Summary

Adds a complete local development workflow for model-engine so developers can iterate on both control plane code and the full endpoint lifecycle without cloud credentials or prod images.

Control-plane-only mode (make dev-server):

• Spins up Postgres + Redis via docker-compose
• LOCAL=true activates fake queue/docker/k8s implementations (mirrors CIRCLECI=true)
• Full gateway API available at :5000 with auth skipped — no k8s cluster needed

Full end-to-end mode (make dev-server-full + make dev-service-builder + make dev-k8s-cacher):

• make kind-up + make kind-image creates a local kind cluster and loads model-engine:local into it
• Service Builder picks up endpoint creation tasks from local Redis and creates real k8s Deployments in kind
• K8s Cacher polls kind and writes endpoint status back to Redis
• Echo server (model-engine:local) used as the inference container — no GPU required

Code fixes included:

• service_builder/celery.py + celery_task_queue_gateway.py: onprem cloud provider now uses redis Celery backend instead of s3 — without this, the Service Builder writes results to Redis but the Gateway looks in S3, leaving endpoints stuck in PENDING
• dependencies.py: LOCAL=true + cloud_provider=onprem falls through to real OnPremQueueEndpointResourceDelegate instead of the fake
• env_vars.py: GIT_TAG defaults to "local" when LOCAL=true so k8s templates reference the correct model-engine:local image

New files:

• docker-compose.local.yml — Postgres 15 + Redis 7 with healthchecks and persistent volume
• service_configs/service_config_local.yaml — HMI config for local services
• model_engine_server/core/configs/local-full.yaml — onprem infra config for kind
• Makefile — all dev targets in one place

Test plan

• make dev-up && make dev-migrate && make dev-server — gateway starts, GET /v1/model-endpoints returns 200
• make kind-up && make kind-image — kind cluster created, model-engine:local loaded
• make dev-server-full + make dev-service-builder + make dev-k8s-cacher — all three processes start cleanly
• POST a sync CPU endpoint with the echo server image → pod appears in kubectl --context kind-llm-engine get pods -n model-engine and endpoint transitions to READY
• Existing unit tests pass: make test

Closes MLI-6681

🤖 Generated with Claude Code

Greptile Summary

• Introduces a complete local dev workflow: make dev-server (control-plane-only with fake k8s/queue/docker) and make dev-server-full + make dev-service-builder + make dev-k8s-cacher (full end-to-end via kind), backed by Postgres 15 + Redis 7 via docker-compose.
• Fixes the onprem Celery backend mismatch: both celery_task_queue_gateway.py and service_builder/celery.py now select redis as backend_protocol for onprem (previously s3), consistent with dependencies.py which already routed onprem to Redis task queues.
• LOCAL=true now defaults GIT_TAG to \"local\" (avoiding the missing-tag error), routes the gateway to the real OnPremQueueEndpointResourceDelegate when cloud_provider=onprem, and uses Redis task-queue gateways instead of SQS.

Confidence Score: 5/5

Safe to merge — all previously flagged P1s are addressed and no new issues found.

All three prior P1 findings (cache_redis_aws_url, ML_INFRA_SERVICES_CONFIG_PATH not pinned, celery_task_queue_gateway.py backend mismatch) are resolved in the current diff. The backend_protocol fix is applied symmetrically in both the Gateway and Service Builder. local-full.yaml sets celery_broker_type_redis: true, keeping broker and backend consistent for the onprem/kind path. No new logic errors identified.

No files require special attention.

Important Files Changed

Filename	Overview
model-engine/model_engine_server/infra/gateways/celery_task_queue_gateway.py	Adds onprem to the redis backend_protocol condition, fixing the Gateway/Service Builder result-backend mismatch for onprem/local deployments.
model-engine/model_engine_server/service_builder/celery.py	Mirrors the celery_task_queue_gateway.py fix — adds onprem to the redis backend_protocol tuple so Service Builder and Gateway agree on where Celery results are stored.
model-engine/model_engine_server/api/dependencies.py	Two targeted changes: LOCAL && !onprem falls through to FakeQueueDelegate; LOCAL is added to the Redis task-queue branch so the control-plane server connects to local Redis instead of SQS.
model-engine/Makefile	New dev-workflow Makefile; ML_INFRA_SERVICES_CONFIG_PATH is now explicitly pinned in both LOCAL_ENV and FULL_LOCAL_ENV, and dev-up delegates readiness polling to docker compose --wait which has its own timeout.
model-engine/service_configs/service_config_local.yaml	Uses cache_redis_onprem_url (not cache_redis_aws_url), correctly bypassing cloud-provider assertions; dummy SQS/billing values are safe for local use.
model-engine/model_engine_server/core/configs/local-full.yaml	New onprem/kind infra config; celery_broker_type_redis: true ensures Service Builder uses Redis broker, consistent with the gateway's Redis task-queue path for onprem.

_{Reviews (9): Last reviewed commit: "docs(devx): replace curl endpoint exampl..." | Re-trigger Greptile}

[lilyz-ai]

@greptile review

[lilyz-ai]

/greptile

[lilyz-ai]

@greptile review

[lilyz-ai]

/greptile