01 / 08 AIESCU LLC · production AI, shipped or taught
Production AI, built for your business.
Work directly with an engineer to ship RAG, agents, and document pipelines. Start with a free scoping week, then a four-week build for a flat $20k. Prefer to build it yourself? Start with the free skills workflow.
curl -fsSL https://raw.githubusercontent.com/aiescu/agent-skills/main/install.sh | bash -s -- --agent claude-codeThe 10 most-starred SKILL.md packs, each installed from its original repo. Also works for copilot, opencode, windsurf, antigravity, kiro and amp.
02 / 08 Proof · client production systems
Proof
- 99.9%
- API availability on a client's production platform
- 10×
- cut in a client's infrastructure costs
- 93%
- retrieval accuracy on a client's production document AI (CI-gated)
- 15+
- years shipping production software
- RAG + agent stack live in 4 weeks
- Eval CI gates every PR
- Observability from day one
- Brief open for inspection
03 / 08 Done for you · 4-week sprint · flat price
We build production AI for clients. It's the living proof of everything we teach.
Work with us on the whole system or one component of it. Every component ships with the versioned skill our agents loaded to build it.
Chunking, hybrid retrieval (pgvector + OpenSearch), recall measured in CI — not assumed.
Federated + semantic search, reranking, MCP-compliant search surfaces.
Airflow / Temporal orchestration — idempotent, schema-versioned, resumable.
Tool-calling loops with guardrails, MCP tool servers, scoped and audit-logged write actions.
Golden sets and judge cassettes that gate every merge — regressions stop before production.
OCR, parsing, embedding, index sync — documents into queryable knowledge.
OTel traces, per-tenant token cost accounting, alerting that fires before customers notice.
Containerized, autoscaled, circuit-broken — operated under real traffic.
$20k flat · per 4-week sprint
No hourly billing, no range. You know the number before the build starts. Larger scopes get a fixed multi-sprint quote on the scoping call.
- A free scoping week first: discovery + spec, deliverable agreed up front
- RAG, agents, ingestion and eval CI behind your data
- Runs inside your AWS account when data is regulated
- You own the code, ADRs, spec and eval suite. No retainer
Choose Done-for-you if…
- You need it in production fast, without hiring an AI team
- You'd rather own the finished system than build it
- You have real data and a real problem right now
Choose Do-it-yourself if…
- You're an engineer who wants to build it yourself
- You want the skills for the long term, not just a deliverable
- You prefer self-paced, with lifetime access
04 / 08 Process · scoping call to handover
Five weeks from first call to a system you own.
Scoping call
Bring a real problem and a sample of your data. We walk you through a working POC and you leave with a fixed number.
free · 30 min
Discovery + spec
Spec hard, code soft. We agree the deliverable and write the chunked implementation spec and ADRs before anything gets built.
free · week 1
Build sprint
Pipeline, agents and eval CI behind your data. Every PR passes the eval gate. If something slips we cut scope, not quality. Either side can stop after the first sprint week for a flat 25% of the fee.
weeks 2–5 · $20k flat
Handover
Code, ADRs, spec, eval suite and the relevant skills transfer to your team, so you keep shipping after the sprint.
no lock-in
05 / 08 Agent skills · source aiescu/agent-skills · 2026-09-04
The 10 most popular agent skills, one install away.
The most-starred SKILL.md packs for Claude Code, Codex, Gemini CLI, Cursor and 70+ other coding agents. Nothing is copied: every skill installs from its original repository and keeps its author and license.
| Rank | Pack | Stars | Skills | Flagship | Install |
|---|---|---|---|---|---|
| 1 | obra/superpowersAn opinionated software-development methodology delivered as composable skills: brainstorm, plan, TDD, debug, review, ship. Session hooks make the agent actually use them. | 281.7k | 14 | brainstorming | npx skills@1 add obra/superpowers |
| 2 | mattpocock/skillsSmall, composable engineering skills straight from Matt's .agents directory. Explicitly against process-owning frameworks. | 249.4k | 37 | grill-with-docs | npx skills@1 add mattpocock/skills |
| 3 | multica-ai/andrej-karpathy-skillsA single skill distilled from Karpathy's observations on LLM coding pitfalls: think first, simplicity, surgical changes, goal-driven execution. | 210.1k | 1 | karpathy-guidelines | npx skills@1 add multica-ai/andrej-karpathy-skills |
| 4 | anthropics/skillsThe official reference collection, the Agent Skills spec, and the skill template. Home of skill-creator and mcp-builder. | 174.0k | 19 | skill-creator | npx skills@1 add anthropics/skills |
| 5 | nextlevelbuilder/ui-ux-pro-max-skillDesign intelligence for building professional UI across platforms: 192 reasoning rules, 79 searchable styles, a design-system generator. | 124.9k | 7 | ui-ux-pro-max | npx skills@1 add nextlevelbuilder/ui-ux-pro-max-skill |
| 6 | JuliusBrussee/cavemanCuts about 65% of output tokens by making the agent talk like a caveman while keeping technical accuracy. Plus subagent presets and memory-file compression. | 103.4k | 20 | caveman | npx skills@1 add JuliusBrussee/caveman |
| 7 | addyosmani/agent-skillsProduction-grade engineering skills organized as a lifecycle: define, plan, build, verify, review, ship. Includes slash commands and agent personas. | 92.2k | 25 | spec-driven-development | npx skills@1 add addyosmani/agent-skills |
| 8 | Leonxlnx/taste-skillGives your AI good taste: stops generic frontend slop with brief inference, variance/motion/density dials, and a design-system map. | 84.3k | 13 | design-taste-frontend | npx skills@1 add Leonxlnx/taste-skill |
| 9 | ComposioHQ/awesome-claude-skillsA curated awesome-list of about 180 skills plus 800+ app-automation skills for Composio's MCP gateway. | 74.5k | 865 | connect-apps | link-only, see repo |
| 10 | ayghri/i-have-adhdTen output rules that stop the agent burying the answer: lead with the next action, number steps, cap lists at five, no preamble. | 27.1k | 1 | i-have-adhd | npx skills@1 add ayghri/i-have-adhd |
06 / 08 Do it yourself · self-paced · lifetime access
Read the map, then do the reps.
Three steps from AI worker to AI entrepreneur. Start with the book. The two programs are the training, and both come with the 300+ skill library.
The AI Entrepreneur
$39.99
launch-price reservation · $99.99 list at launch
Build with Claude Code
$297
founder price, waitlist open · $997 regular
- Ship a working Claude-Code agent end-to-end
- File-edit, MCP tools, hooks, slash commands
- Lifetime access + the skills library
Become an AI Engineer
$499
founder price, waitlist open · $2997 regular
- RAG, agents, evals, observability, prod-deploy, multi-agent
- Ship a full RAG + agent + eval-CI stack
- Lifetime access + updates + skills library
For engineers who can code, founder-engineers and eng-led teams of one. Not for "make me an AI app without writing code".
07 / 08 Laws we live by · LLMs in production
10 laws for shipping AI to production without burning cash
What we've learned building with LLMs. Every build and every course follows them.
- 01
Spec hard, code soft.
A page of working spec is worth a week of throwaway code. LLMs accelerate the wrong thing if the spec is wrong.
- 02
Evals before the model call.
Write the failing eval first. Without a passing bar you don't have a product, you have a forever-prototype.
- 03
Tools beat prompts.
A 20-line tool with a strict schema beats a 2,000-token system prompt. The model recovers from a wrong tool call; it doesn't recover from a vague instruction.
- 04
Cache aggressively, route ruthlessly.
Prompt cache, embedding cache, response cache. Cheap model routes; expensive model produces. 80% of the bill is the wrong model on the wrong call.
- 05
Monolith for LLMs, microservices for humans.
LLMs read a monolith faster than a 12-service mesh. Split early and you lose the one thing they're best at: holding the whole system at once.
- 06
Curate the context, don't polish the prompt.
The model is only as good as what's in the window. Cut every token that isn't a fact, an example, or a constraint.
- 07
Receipts, not claims.
Every “it works” is a test run, an eval row, or a git log line. Vibes don't ship.
- 08
Subagent review = 2 stages.
Finder + adversarial verifier. One agent gets confidently wrong; two disagreeing agents force the truth.
- 09
Manager mode: 5 → 1.
One operator orchestrating 5 parallel agents ships what used to need a team. Lanes, async, documented handoffs.
- 10
Player-coach.
Ship code and review agents in the same day. The skill is reading whether today needs hands-on-keyboard or orchestration.
08 / 08 Common questions
Asked on almost every call.
How much does it cost?
A flat $20k per 4-week build sprint; the scoping call and the scoping week before it are free. No hourly billing, no range — you know the number before the build starts. Larger scopes get a fixed multi-sprint quote on the free scoping call.
What if you miss the 4-week timeline?
That's what the scoping week is for. We agree the deliverable up front, and if something slips we cut scope, not quality. You ship something that works.
Is my data safe? Can it stay in my cloud?
Yes. For regulated or proprietary data we run embeddings and inference inside your own AWS account (Bedrock / OpenSearch). Write actions are scoped, confirmed, and audit-logged.
Do I own the code?
Completely. You get the code, ADRs, a chunked implementation spec, and eval CI — transferred to you. No lock-in, no mandatory retainer.
Is this cohort-based?
No. It's self-paced with lifetime access. Start and finish whenever.
Can I see proof before I commit?
Yes. There are live demos, the numbers above, and a working POC I walk you through on the call. The full brief with NDA-gated receipts is available too.
Bring a real problem and a sample of your data.
One scoping call. You leave with a fixed price and a four-week plan, or a pointer to the program that gets you there yourself.