AI systems that ship in production
Hiring rounds often take weeks before anyone writes real code. What you usually need is one person who can own the architecture and the delivery from the first week.
We start with your constraints and what success looks like, then build in short demos so everyone can react to working software instead of documents.
If that sounds like your situation, Arcode is set up for startups and product teams hiring AI engineering talent. Agents, MCP servers, RAG pipelines, and LLM features wired into real software rather than demos.
What you get
- Production AI agents with tool calling
- MCP servers connected to your stack
- RAG over company knowledge
- OpenAI and Claude product integrations
- One studio accountable for the result
- Design and engineering on the same schedule
- Support available after launch
How engagement works
As your AI engineer, we run a shared backlog, design reviews, and engineering sprints, and you keep one point of contact throughout.
- 01
Discover
Goals, constraints, data sources, and success metrics, all settled before we touch architecture or prompts. You see progress on the critical path every week.
- 02
Design
System design, AI/tool boundaries, interface direction, and interaction prototypes. Nothing here waits on a handover between separate teams.
- 03
Build
Modern stack, clean architecture, evaluated AI behavior, and iterative demos you can react to. On AI engineering projects we favour practical architecture over trendy defaults.
- 04
Launch
Ship, monitor, harden, and keep improving models, tools, and product after go-live. The work for teams in Berlin stays tied to the outcome you asked for.
How we build it, and why
The stack is chosen per project, not applied from a template. For AI engineer work these are the defaults, and the reason each one is on the list.
- OpenAI
- a second model provider, so a single vendor's outage or price change is not your problem
- Claude
- long-context reasoning and reliable tool calling for agent work
- MCP
- the Model Context Protocol, so your tools plug into Claude, Cursor, and other agent clients without a custom adapter for each
- LangChain
- orchestration for multi-step chains when the flow is genuinely complex
- Next.js
- server rendering and static generation in one framework, so marketing pages stay fast and app routes stay dynamic
- NestJS
- structure and dependency injection on the backend, which keeps a growing API from turning into a pile of route handlers
What the first weeks look like
- Days 1-3
Context and constraints
Codebase, deploy path, and what "done" means for the AI engineering. Short, because the useful version of this is specific.
- Week 1
First shipped slice
Something real goes out in the first week — production ai agents with tool calling rather than a setup ticket.
- Weeks 2-6
Build to the milestone
Pull requests, reviews, and weekly demos. You see the work as it happens rather than at a handover meeting.
- Handover
Documented and transferable
Runbooks, architecture notes, and a walkthrough, so the work does not depend on us still being here.
Weighing up your options
There are three realistic ways to get this built. Each is the right answer for someone.
A solo freelancer
Cheapest per hour and fine for a contained task. The risk on AI engineering is breadth — one person covering design, backend, infrastructure, and launch usually means one of them is weak, and there is no cover when they are unavailable.
A local agency in Berlin
Same timezone and a face to meet, which genuinely matters for some teams. You are also paying Berlin startups hiring for shipping speed and modern stacks. rates for the whole team including the layers that never touch your AI engineering, and a local shortlist is a small shortlist.
Arcode
A small senior team. The people who scope the AI engineering write the code, you get weekly demos instead of status decks, and the engagement ends when the thing is live and handed over.
Work we have shipped
Note Assist
AI product · Next.js · Speech
AI meeting assistant with live speech-to-text, ChatGPT chat, and Firebase auth for transcription workflows.
Freelance AI product built with Next.js, Firebase, AssemblyAI, and ChatGPT for live meeting transcription and realtime chat around the transcript.
- · Live speech-to-text transcription
- · Realtime chat over meeting notes
- · Firebase authentication
- · AI-assisted follow-up on transcripts
Next.js · Firebase · AssemblyAI · ChatGPT · TypeScript
Brain CMS
Frontend · Firebase · Analytics
Multi-module enterprise CMS with documents, calendars, video, and advanced analytics.
Content and operations system spanning enterprise modules with Firebase and rich media tooling.
- · Multi-module business domains
- · PDF and document management
- · Advanced charting suites
- · Calendar and media workflows
React · Redux Saga · Firebase · Chart.js · Video.js
CRM Dashboard
Frontend · Analytics
Customer relationship dashboard with management workflows and data visualization on Next.js.
Freelance CRM surface for managing relationships and visualizing key metrics, deployed on Vercel.
- · CRM management views
- · Data visualization
- · Responsive dashboard layout
- · Vercel deployment
Next.js · React · Tailwind · Vercel
Working with teams in Berlin
Berlin startups hiring for shipping speed and modern stacks. We work with clients in Berlin remotely and keep overlapping hours in CET/CEST. We usually collaborate in German/English.
How we work with Berlin teams
Berlin is three to four hours behind us, so there is a full shared afternoon. Berlin teams tend to move fast and decide in writing, which suits the way we run a backlog.
What Berlin clients usually need
Berlin is the most startup-weighted market we work in outside the US: early-stage products that need to be live before the next raise, and AI features added to products that already have traction.
Contracts, data, and compliance
We invoice in EUR with reverse-charge VAT. GDPR is scoped into the build, with EU-hosted data by default and a DPA provided as standard rather than on request.
