Prompt Claude to Plan 3 Production-Grade AI Portfolio Projects
Bashiri Smith · Facebook reel · 2026-10-11 · 1:11 · 18,190 views · Open on Facebook
Topics: Portfolio Projects, AI Dev Tools & Productivity, AI System Design & Architecture · Level: intermediate
Summary
Bashiri Smith shares a prompt to paste into Claude before you start any coding project. It asks Claude to act as a principal engineer and design three AI-native portfolio projects. Each project comes with an 8-step Claude Code pairing plan: scaffolding, API contracts, data, workers, tests, AI (RAG or agents plus evals), CI/CD and instrumentation. The prompt adds rules that rule out toy apps: each project must handle real-world issues, set measurable eval, latency and cost targets, and use a different domain from the others.
Key points
- Prompt framing: 'You are a principal engineer tasked with designing three AI-native portfolio projects', with a Claude Code pairing plan of 8 concrete, paste-ready prompts per project.
- Steps 1-2: Scaffolding (a monorepo file structure for your chosen stack) and Contracts (OpenAPI spec and GraphQL schemas, plus generated typed clients).
- Steps 3-4: Data (Postgres schemas, migrations, seed scripts, fake data) and Workers (a queue with one job that has retries and dead-letter handling).
- Steps 5-6: Tests (Jest/PyTest setup with example unit and integration tests) and AI (RAG or agent loops, an evaluation harness and a small golden dataset).
- Steps 7-8: CI/CD (GitHub Actions for test, build and Docker push) and Instrumentation (OpenTelemetry).
- Constraint: no toy apps. Each project must deal with backpressure, retries, pagination, partial failures and schema migrations.
- Use the TypeScript or Python stacks most teams use today. Avoid vague AI claims and set precise evals plus latency and cost targets.
- Each of the 3 projects should use a different domain and problem type to show range.
Resources mentioned
- Claude Opus 5.5 · tool · claude.ai · paid · open in a browser to verify · recommended by both Bashiri Smith & Melvin Vivas
Anthropic's AI assistant, used throughout the guide to tailor resumes, add live roles to the tracker, match connections to target companies and find hiring managers.
Also in: Claude models (Opus 5.5) generally available on Google Cloud (Melvin Vivas on X · notes), Developers moving from ChatGPT/Codex to Claude (Melvin Vivas on X · notes), Claude Duet: Anthropic's Research Preview of a Proactive Agent App (Melvin Vivas on X · notes), Using Claude Opus 5.5 to Produce Short-Form Videos in About 5 Minutes (Melvin Vivas on X · notes) and 81 more - Claude Code · tool · code.claude.com · paid · recommended by both Bashiri Smith & Melvin Vivas
Build agents and pipelines from the terminal; the guide's main agentic coding tool.
Also in: First impressions: Claude Code in the cloud is good and fast (Melvin Vivas on X · notes), Why Claude Code feels faster: gzipped turns and instant cloud sessions (Melvin Vivas on X · notes), Watch What Fills Your Coding Agent's Context and When to Clear (Melvin Vivas on X · notes), Claude Code's Right Sidebar and Context Breakdown (Melvin Vivas on X · notes) and 124 more - AI Projects (Google Drive folder) · pdf · drive.google.com · free
The creator's guide to 27 AI projects, with full architecture, build steps phase by phase, interview talking points, and the stack for each project.
Also in: 5 AI Engineer Skills Mapped to 5 Portfolio Projects (Bashiri Smith on Facebook · notes), 25 AI Engineer Interview Questions Covering ~95% of Interview Topics (Bashiri Smith on Facebook · notes), 6-Step Framework for Building AI Projects Target Companies Care About (Bashiri Smith on Facebook · notes), Generate Job-Worthy AI Project Ideas from Recent Research Papers + ChatGPT (Bashiri Smith on Facebook · notes) and 6 more - OpenAPI Specification · docs · openapis.org · free
The standard format for describing REST APIs, which can be used to generate typed clients. - GraphQL · tool · graphql.org · free
A query language and schema system for APIs. - PostgreSQL · tool · postgresql.org · free · recommended by both Bashiri Smith & Melvin Vivas
An open-source relational database; with the pgvector extension it also stores and searches embeddings.
Also in: Pi Durable: A Runtime for Crash-Proof, Long-Running, Multiplayer Agents (Melvin Vivas on X · notes), Why Plan Mode Still Matters in AI Coding Agents (Codex, Claude, Cursor) (Melvin Vivas on X · notes) - Jest · tool · jestjs.io · free
A JavaScript/TypeScript testing framework. - pytest · tool · docs.pytest.org · free
Python's most used test framework: plain assert statements, fixtures and a simple test runner. - GitHub Actions · tool · github.com · free · recommended by both Bashiri Smith & Melvin Vivas
GitHub's built-in CI/CD: workflows that run tests, builds and deploys on every push.
Also in: 5 Weekend AI Engineering Projects: Cost Routing, Caching, Evals & Observability (Bashiri Smith on Facebook · notes), Use Claude Code /loop to Watch and Auto-Fix Failing GitHub Actions (Melvin Vivas on X · notes) - Docker · tool · docs.docker.com · free · recommended by both Bashiri Smith & Melvin Vivas
Containerize everything you ship.
Also in: Install Docker on your agent's computer to run anything (Melvin Vivas on X · notes), Deploying AI workflow integrations with Apache Camel in Docker (Melvin Vivas on X · notes), Devin's cloud Ubuntu sandbox ships with Docker pre-installed (Melvin Vivas on X · notes), 7 Habits to Become an AI Engineer: Books, Tooling, Research & Shipping (Bashiri Smith on Facebook · notes) and 7 more - OpenTelemetry · tool · opentelemetry.io · free
Vendor-neutral tracing standard. The guide recommends it, with its GenAI semantic conventions, as the base layer for agent observability.
Also in: 7 Habits to Become an AI Engineer: Books, Tooling, Research & Shipping (Bashiri Smith on Facebook · notes) - TypeScript · tool · typescriptlang.org · free · recommended by both Bashiri Smith & Melvin Vivas
A typed superset of JavaScript.
Also in: One-Shotting a Slack Clone with Fable 5 (Melvin Vivas on X · notes) - Python · tool · python.org · free · recommended by both Bashiri Smith & Melvin Vivas
The programming language most AI engineering work is done in; python.org has the installers and the official tutorial.
Also in: Step-by-Step Roadmap to a $200K+ AI Engineering Role (Bashiri Smith on Facebook · notes), Software Engineer to AI Engineer: Job Boards, Stack, Projects and Learning Sites (Bashiri Smith on Facebook · notes), Run Codex in the Cloud: Spin Up DigitalOcean Droplets with the Codex Plugin (Melvin Vivas on X · notes)
Try this
- Before starting any coding project, paste the principal-engineer prompt into Claude to design 3 AI-native portfolio projects.
- Ask for 8 paste-ready Claude Code prompts per project: scaffolding, contracts, data, workers, tests, AI, CI/CD and instrumentation.
- Add the constraints: no toy apps; handle backpressure, retries, pagination, partial failures and schema migrations.
- Set precise evals and latency and cost targets, and build a small golden dataset.
- Pick a TypeScript or Python stack and a different domain for each project.
- Optionally, comment 'Claude' on the reel to get the full prompt and 27 project ideas.
- Three AI-native portfolio projects in different domains, each built as a monorepo with OpenAPI/GraphQL contracts and typed clients, Postgres schemas and migrations, a queue worker with retries and dead-letter handling, Jest/PyTest tests, a RAG or agent loop with an eval harness and golden dataset, GitHub Actions CI/CD with Docker push, and OpenTelemetry instrumentation.
More in Portfolio Projects
- Opus 5.5 showcase: a wildfire spread analysis prototype
- Use your own product before you ship it
- ChatGPT Prompt to Generate Standout Portfolio Project Ideas
- Creator shares his GitHub profile (donvito)
- Local Two-Person Video Face Swap: InsightFace + GFPGAN + SegFormer
- Generate Portfolio AI Project Ideas with Claude and Hugging Face Tasks