Skip to content

TARGETS_ROADMAP.md

Latest commit

 

History

History
130 lines (96 loc) · 5.72 KB

TARGETS_ROADMAP.md

File metadata and controls

130 lines (96 loc) · 5.72 KB
Raw

AI Native Lang (AINL) targets roadmap — real-world and production

Expanding targets so AI Native Lang is usable in production and for mass adoption. Major runtime/data/reactive lanes below are now production-ready, with a short future list retained for optional expansion.

Tier 1: Today (implemented)

Target Output Use case
React (browser) JSX, hash router, layout, forms, tables, events Dashboards, SPAs
FastAPI + runtime Python server + ir.json + adapters API backend
Unified web server API + static mount Single process for API + frontend
Prisma schema.prisma DB schema from D
MT5 .mq5 stub Trading bot skeleton
Scraper Python requests + BeautifulSoup Scraping jobs
Cron Python stubs Scheduled jobs
Pay / Cache / Queue Adapters (Stripe, Redis, Bull) Runtime only

Tier 2: Next (high priority for adoption)

Target Purpose Ops / IR
OpenAPI 3.0 API docs, codegen, Postman, gateway config E, D → paths + schemas
Docker + Compose One-command run and deploy S, E, D → Dockerfile + compose
Next.js API routes Serverless API + React on Vercel E, L, R, J → /pages/api/* + React
Vue / Svelte Alternative frontend frameworks Same IR as React (fe, Rt, Lay, etc.)
SQL migrations Create/alter tables from D D → SQL (Postgres/MySQL)
Env + config 12-factor env schema from S/C/P S, C, P → .env.example, config schema

Tier 3: Production (deploy, observe, scale)

Target Purpose Notes
Health + readiness /health, /ready in emitted server From S/core
Structured logging Request/response and label logs ✅ LoggingMiddleware in emit_server
Auth middleware JWT or API-key from AINL (A op) ✅ A op → Depends in emit_server
Rate limit Per-client RPM via env ✅ RateLimitMiddleware; RATE_LIMIT env
Kubernetes Deploy manifest (Deployment, Service, Ingress) ✅ emit_k8s() → k8s.yaml
Native async runtime loop Optional async graph/step execution + async adapter path ✅ AINL_RUNTIME_ASYNC=1 / --runtime-async; redis has full async verb parity, dynamodb supports bounded async streams, and supabase has advanced lightweight fanout/replay/cursor helpers; see docs/runtime/ASYNC_RUNTIME.md
Terraform/Pulumi DB, queue, cache resources 🔲 Planned
CI (GitHub Actions) Test + build + emit from .lang ✅ .github/workflows/ci.yml

✅ Production Ready: Database Adapters

Runtime-native adapters with established contracts and docs:

  • sqlite
  • postgres
  • mysql
  • redis
  • dynamodb
  • airtable
  • supabase

These cover local SQL, cloud SQL, cache/pub-sub, managed NoSQL streams, and realtime SaaS workflows.

✅ Production Ready: Native Async Runtime

AINL's native async runtime loop is production-ready for graph execution with async-capable adapters:

  • async execution path via AINL_RUNTIME_ASYNC=1 or --runtime-async
  • bounded async reactive polling patterns for streams/realtime/pub-sub
  • optional observability integration (including JSONL sink) for production diagnostics

See docs/runtime/ASYNC_RUNTIME.md for runtime behavior and adapter coverage details.

✅ Production Ready: Reactive / Event-Driven Features

Reactive adapter primitives are production-ready:

  • DynamoDB Streams: streams.subscribe, streams.ack, checkpoint helpers
  • Supabase Realtime: realtime.subscribe, realtime.replay, realtime.get_cursor, realtime.ack, fan-out helpers
  • Redis pub/sub: async parity and bounded subscribe flows
  • Durability guidance + production starters:
    • docs/reactive/ADVANCED_DURABILITY.md
    • templates/durability/
    • templates/production/

Tier 4: Ecosystem (other languages and platforms)

Target Purpose
Node/Express or Nest TypeScript backend; same IR
Java (Spring Boot / Quarkus) Enterprise backend
.NET (Minimal API) C# backend
Go (Chi / Echo) High-performance API
GraphQL Schema + resolvers from D, E
gRPC Service def from D, E
React Native / Flutter Mobile from same fe IR
System scripts Shell or Python from Cr, R
Minecraft / game plugins Event hooks from S, E, U

Implementation status

Target Status Emitter / artifact
OpenAPI 3.0 ✅ Implemented emit_openapi() → openapi.json
Docker + Compose ✅ Implemented emit_dockerfile(), emit_docker_compose()
Next.js API routes ✅ Implemented emit_next_api_routes() → next/pages/api/*.ts + health/ready
Vue / Svelte ✅ Implemented emit_vue_browser() → App.vue, emit_svelte_browser() → App.svelte
SQL migrations ✅ Implemented emit_sql_migrations(ir, dialect) → migrations/001_initial.sql
Env/config ✅ Implemented emit_env_example(ir) → .env.example
Health/readiness ✅ Implemented In emit_server(): /api/health, /api/ready; in OpenAPI
Auth (A op) ✅ Implemented A op → services.auth; emit_server() adds Depends for protected routes
Node/Java/.NET/Go 🔲 Planned New emitters + adapters

Adding OpenAPI and Docker to the compiler gives immediate production value: documented, runnable API and one-command deploy for the current stack.

Future / Nice-to-Have

  • Multi-node/shared durability coordination primitives (beyond process-local helper maps).
  • Runtime scheduling optimizations for very high-throughput reactive pipelines.
  • Optional advanced adapter features for specialized deployments (for example deeper provider-specific admin helpers).