Senior Backend Developer

Time: Full-time

Location: Aarhus

Start date: ASAP

Where we are going

Google still matters. But the way people find things online is changing.

Someone researching a purchase might start in ChatGPT, compare options in Perplexity, run a query on Google, then go back to an AI assistant to decide. For companies, that creates a new problem: they have very little idea where they show up, what gets said about them, or which sources influence the answer. Almost none of them can measure it today.

We are building the infrastructure to measure it. The hard part is not the reporting. It is the collection. Every surface we measure is a moving target, and none of them ships a feed for us to consume. So we collect all of it ourselves, and we buy none of it.

Three products, one data layer

AccuRanker is the rank tracking and SEO analytics platform: rankings, competitor movement, keyword research, dashboards, forecasting and much more.
AccuLLM tracks brand visibility inside ChatGPT, Perplexity, AI Overview and AI Mode.
AccuSERP is a search data API that we sell to other companies in the industry.

Here is a normal day at AccuRanker

27.4 billion wide rows inserted, roughly 10 trillion a year.
85 million web requests collected.
80.5 million background jobs run successfully across our worker fleet.
64.2 million analytical queries served.

What you will work on

Scale the collection and ingestion pipeline behind those numbers. Drive down cost and latency per request as volume grows.
Add new search surfaces as they appear, and keep the ones we already cover working when they change shape.
Design storage and query paths for very large time series data, across columnar and relational stores.
Own services end to end, from design through deployment to how they behave in production.

What you bring

Must have:

Distributed systems fundamentals. Queues, workers, backpressure, etc.
Comfort in Linux and close to production. Be good at debugging incidents, understanding system behavior, improving observability, and making services more reliable.
Ownership and initiative. You take a problem, you decide how to solve it, and you follow through.
Curiosity on new tools and technologies.

Good to have:

The network layer in depth. Connection lifecycle, timeouts and retries, TLS, HTTP/2 and HTTP/3
ClickHouse or another columnar store at scale.
AI agents in production. Tool and context design, structured output and evals.
Large scale web scraping or browser automation.
Python and popular frameworks (Django, Celery, pandas/polars, etc.). Our stack primarily runs on Python, but it is not a must if you check off the boxes on strong fundamentals.
Experience with Kafka
Data infrastructure at scale. Experience with the ingestion, storage, and query-performance problems that our scale implies.

Who does well here

We want people who set a high standard for their own work . You care about performance, reliability, cost and operational simplicity.

Four things matter as much as the technical list above.

1. You like owning things

You own a ticket/feature end-to-end, from idea to production. You can make progress while the problem is still half defined. A month after it ships, you are still the one who cares about it.

2. You are pragmatic

You move fast when speed is what the problem needs. You also know when stability, safety and maintainability matter more.

3. You are curious

This field moves fast right now. You read, you try things, and you form your own view of what is worth using. You do not wait to be told what is new.

4. You want a wide job

In one week you might tune a database, debug a proxy network, and design an agent.

This role is not for you if you want a roadmap handed to you, a narrow technical scope, or a large team around you. We are relatively small, the problems change, and you will often be the person deciding both what to build and what “done” means.

AI at AccuRanker

AI is a big part of the product and part of how we build it. AccuLLM sends queries to LLMs all day, customers use our MCP server, and agents and tool design are real engineering problems here.

We do not use agents for the sake of using agents, but it is clear by now that being effective with coding agents is becoming a core engineering skill, and it is a key part of the job here. You should be good at giving agents the right context, breaking work into pieces they can execute well, reviewing what they produce, and knowing when to take over yourself.

The bar is not “uses AI a lot.” It is getting more high-quality engineering done without outsourcing your understanding and judgment.

How we work

We are a small engineering team of 6. Everyone knows what everyone else is building.
You own services, not tickets. You decide how to solve the problem, make the tradeoffs, and deploy it yourself.
We ship many times a day. A change can go from idea to production the same day.
We argue about designs, in writing and in the room. Correctness, cost and latency are worth arguing about out loud.
We share our knowledge and help each other own unfamiliar parts of the system.

What we offer

Salary set individually on skills and experience.
Pension, health insurance, flexible working hours, and paid parental leave.
Hardware of your choosing.
Office in central Aarhus on Åboulevarden, top floors, panoramic views over the city. Great lunch.
On-site.