Specialized training · Anthropic's Claude

Get your team productive
with Claude — fast.

Hands-on Claude training for engineering, operations, and leadership teams. We teach the patterns we use ourselves — on the stack you actually ship with — so your team walks out building real things, not posting demo screenshots in #general.

Book a curriculum call See the curriculum

Remote, onsite, or hybrid · Custom curricula · 1-on-1 through team-of-50

Why this matters

The model has gotten ahead of most teams.

The companies winning with AI right now aren't the ones with the most engineers — they're the ones whose people actually know how to work with these tools. The skills are real, learnable, and unevenly distributed.

Most AI training in the wild is one of two things: a vendor pitch that's really about their platform, or a generic "prompt engineering" course that won't survive contact with a real production system. Neither closes the gap.

We do the engineering work ourselves. We use Claude every day to ship customer software. The training we deliver is the institutional knowledge of a practicing AI engineering shop — handed to your team in days, not quarters.

What you'll learn

Three tracks, taught by people who do this work.

We tailor each engagement to your stack, your team, and the problems you're trying to solve. The tracks below are the most common shapes — pick one, mix them, or design something custom with us on a curriculum call.

Track 1

Engineering

For software teams integrating Claude into products, internal tools, and developer workflows. Hands-on with the API and the SDKs your team actually ships with.

//

Foundations: the mental model

How Claude works at a level deep enough to build on. Model family (Opus, Sonnet, Haiku) and when to use each. Context windows, tokens, and the things that quietly cost money.

//

Prompt patterns that survive in production

System prompts vs. user messages. Structured output with XML tags. Chain-of-thought, few-shot, and self-critique. When fine-tuning is the wrong answer and when it isn't.

//

Tool use, function calling, and agents

Giving Claude the right tools and the right rails. When to let Claude drive vs. when to keep the orchestration on your side. Agent loops, hand-off rules, and what to never automate.

//

Claude Code as a daily driver

Working alongside Claude in your codebase. Slash commands, skills, hooks, MCP servers, and the workflow patterns that actually save hours per day.

//

Cost, latency, observability

Prompt caching, batch requests, streaming, and model routing. How to measure what your AI features actually cost — and what to do when the bill surprises you.

//

Hardening and failure modes

Prompt injection, retries, fallbacks, model migrations. Building features that degrade gracefully when the model has an off day.

Track 2

Operations

For operations, customer success, and internal-tools teams adopting AI-assisted workflows. Less code, more about getting real work done with Claude in the loop.

//

Where Claude fits in your existing tools

Slack, Teams, Notion, internal portals. Practical patterns for embedding Claude where work already happens — without rewiring your whole stack.

//

Document workflows

Extracting structured data from contracts, invoices, reports, and email. Summarization that holds up. Classification and routing that doesn't need re-checking every Monday.

//

Customer-facing AI — when and how

PII handling, escalation paths, conversational guardrails, and the difference between "Claude answers" and "Claude drafts, a human sends."

//

Cost governance for non-engineers

How to read a Claude usage report. Setting up alerts. Knowing when to ask engineering to switch models.

Track 3

Leadership

For executives, founders, and team leads building an AI strategy. What's real, what's hype, and how to move on the opportunities your competitors are still studying.

//

What's actually shippable today

A grounded view of what Claude can do well in production, what's still rough, and what's coming soon. No demos that won't survive contact with real users.

//

Build vs. buy vs. wait

A framework for deciding which AI features to build in-house, which to buy from SaaS, and which to deliberately not pursue yet. With examples from real engagements.

//

Team enablement

Getting your org to actually use the tools. Training paths by role, tool selection, and the cultural patterns that turn AI from a curiosity into a habit.

//

Risk: security, compliance, vendor lock-in

Prompt injection, data leakage, model drift, vendor concentration. Where the real exposure is — and where the audit checklists overreach.

Format & delivery

The shape that fits your team.

Every engagement scopes around what your team actually needs. The four formats below cover most situations — anything else, we build for you.

Half-day intensive

A focused starter

Foundations plus one track. The shortest path to a team that knows what's real, what's worth trying, and what to do next. Best for teams just starting with Claude.

Up to 25 people · Remote or onsite

Full-day workshop

Hands-on with your stack

Foundations plus two tracks, with hands-on lab time in your actual codebase or workflow. The team writes real prompts and builds real integrations under our coaching.

Up to 20 people · Remote or onsite

Multi-day deep dive

Build something real, together

Two to five days. Foundations, all three tracks, and a guided build of a real feature your team takes home and ships. The most common shape for serious adoption.

Up to 15 people · Onsite preferred

Ongoing coaching

Embedded for the rollout

Weekly or bi-weekly engagement with your team as you roll Claude into more of your work. Office hours, design reviews, on-call for tricky decisions.

3 / 6 / 12-month engagements · Remote

Who this is for

Teams ready to do real work with Claude.

Engineering teams

You're integrating LLMs into products or internal tools and want the patterns that hold up at scale. You're comfortable with code, less comfortable with the gap between "demo works" and "production-grade."

Operations & internal-tools teams

You've seen what Claude can do and you want to put it to work in your actual processes — not as a demo, but as a habit. You need patterns that don't require a developer to babysit.

Leadership building AI strategy

You've heard the pitches. You want a grounded view of what's worth building, what's a distraction, and how to enable your team without burning a quarter on consultants who don't ship code.

What you walk away with

Outcomes, not certificates.

Every engagement is designed around what your team can do differently the Monday after we leave. We don't issue a piece of paper — we leave you with capability you can prove.

  • →

    A shared mental model across the team for how Claude works and where it fits in your stack.

  • →

    A working starter codebase or workflow built in your environment during the engagement, that the team owns and can extend.

  • →

    A pattern library of prompts, tool definitions, and agent shapes that work for your domain — not generic examples from a tutorial.

  • →

    A decision framework for what to automate with Claude, what to keep humans on, and what to revisit in six months.

  • →

    An open line to us for the questions you don't know you'll have until you start shipping.

Why 159 Networks

We do this work for a living, every day.

159 Networks is a senior AI engineering shop. We build production systems on top of Claude for customers across industries. This site itself was built largely with Claude in the loop — including the booking flow that scheduled your call, the calendar grid you used, and the Cloudflare Pages Functions behind it.

That means the training we deliver isn't drawn from a slide deck. It's drawn from the engineering decisions we make in real customer code — what works, what fails, what costs too much, what's worth fighting for.

Our trainers are practicing engineers, not full-time instructors. Every session is led by someone who shipped Claude-backed code that week. When your team asks a question that the documentation doesn't answer, we answer it from experience.

And we're an independent shop — not an Anthropic reseller, not a SaaS vendor with a course attached. We have no incentive to push you toward a tool that doesn't fit. We tell you what's actually worth your team's time.

Common questions

FAQ

How long does a typical engagement take? +

Anywhere from a half-day starter to a 12-month embedded engagement. Most customers land in the 2-3 day workshop range to start, often with ongoing coaching after. We scope on the curriculum call.

What does it cost? +

We quote per engagement, not per seat. Pricing depends on scope, team size, format, and whether we're working in your codebase. The curriculum call is free and we'll give you a real number before anyone signs anything.

Do you offer training on other models — GPT, Gemini, open-source? +

Most patterns we teach transfer across frontier models. We focus on Claude because it's what we ship with most, but if your team is on another stack, we can adapt. Ask on the call.

Are you an Anthropic-affiliated training provider? +

No. We're an independent engineering shop that ships on top of Claude in production. We have no commercial relationship with Anthropic — which is part of why our recommendations are unbiased.

Can the training be onsite? +

Yes. US-based onsite is standard. International by quote. Multi-day formats are usually better onsite — the energy is different when you're in a room together.

What if my team is brand-new to AI / LLMs entirely? +

Perfect audience. The Foundations module is built for exactly that — and the Operations and Leadership tracks don't require any coding background. We meet teams where they are.

Next step

Let's design the right curriculum for your team.

Thirty minutes. We'll learn what your team is trying to do with Claude, tell you straight whether training is the right move, and — if it is — sketch the engagement that fits.

Book a free 30-minute curriculum call

Or email training@159.network

Preview B Switch to A