Python · AI · data
Python-developers inhuren of werven, gespecialiseerd in AI
Python People heeft een eigen netwerk van ervaren Python-developers. Daaruit bemiddelen we freelancers voor opdrachten en werven we developers voor een vaste plek in je team. Freelancers op senior- en architectniveau, vaste functies van medior tot architect. Gespecialiseerd in AI, data en backend.
- Binnen 48 uur een eerste selectie, vrijblijvend
- Flexibel opzegbaar
- Geen kosten bij verzuim of vakantie
Een freelance Python-developer voor je opdracht
Nederlandstalige freelancers uit ons eigen netwerk van Python-developers, zelf gesproken en technisch getoetst. Binnen 48 uur heb je een eerste selectie, vrijblijvend, en je weet vooraf wat het uurtarief is.
Geschikt voor: teams die tijdelijk extra kennis of capaciteit nodig hebben. Freelancer inhuren Werving en selectieEen Python-developer in vaste dienst
Uit ons eigen netwerk van Python-developers stellen we kandidaten voor een vaste functie in je team voor, ook mensen die niet actief op zoek zijn. Iedereen is technisch getoetst. 100% No Cure, No Pay, met een garantie in de eerste maand.
Geschikt voor: organisaties die kennis blijvend in huis willen halen. Bekijk werving en selectieBen je Python-developer? Bekijk opdrachten en vacatures
Veelgevraagd in onze opdrachten
- Python 3.12
- FastAPI
- Pydantic
- pytest
- PostgreSQL
- pgvector
- LangGraph
- Airflow
- dbt
- MLflow
- Kubernetes
- Azure OpenAI
Onder meer gewerkt voor
- Adyen
- Brink’s
- Eurofins
- Fox-IT
- GS1
- Insmed
- KPN
- Vacansoleil
Freelancer inhuren
Zo werkt het
Vier stappen, van je eerste vraag tot de eerste werkdag.
- 01
Vertel wat je nodig hebt
Bel of mail ons. Een vacaturetekst is genoeg: wij vertalen die naar het juiste technische profiel.
- 02
Wij stellen kandidaten voor
Binnen 48 uur krijg je passende kandidaten uit ons eigen netwerk van Python-developers.
- 03
Kennismaken, vrijblijvend
Je kiest wie je wilt spreken en voert zelf het gesprek. Zo weet je snel of iemand bij je team past.
- 04
De developer begint
Kies je voor een kandidaat, dan kan die vaak al binnen een week aan de slag.
Werving en selectie
Liever iemand in vaste dienst?
Wij nemen het wervingsproces van je over, van vacature tot eerste werkdag, met aandacht voor je team en je cultuur. 100% No Cure, No Pay.
- 01
Vacature
Je stuurt ons de vacature of functieomschrijving. In een kort gesprek bespreken we je wensen, het team en de cultuur.
- 02
Ons netwerk
Kandidaten uit ons eigen netwerk van Python-developers, ook mensen die niet actief op zoek zijn.
- 03
Selectie
We spreken de kandidaten en toetsen hun technische kennis, zodat je alleen kandidaten ontvangt die volledig matchen.
- 04
Kennismaken en start
Je spreekt de beste kandidaten. Wij begeleiden het aanbod tot de eerste werkdag, met één vast aanspreekpunt.
AI-specialisaties
Waar onze Python-developers het verschil maken
Een AI-toepassing raakt je hele stack: van de data aan de onderkant tot de API, de monitoring en de keuzes daarboven. We selecteren per discipline.
Agentic AI
Agents die taken afronden in je systemen, met duidelijke grenzen en volledige traceerbaarheid.
RAG en kennissystemen
Betrouwbare antwoorden uit je eigen documenten, met bronvermelding en rechten per gebruiker.
LLM-integratie
Taalmodellen als betrouwbaar onderdeel van je bestaande software en processen.
AI-backends met FastAPI
Snelle, getypeerde Python-backends die je software en AI-functies veilig en schaalbaar aanbieden.
Data engineering
Python-pipelines die je data schoon, actueel en veilig beschikbaar maken, als fundament onder elke toepassing.
MLOps en LLMOps
Evaluatie, uitrol en monitoring, zodat AI na de lancering blijft presteren.
AI-architectuur
Senior architecten die AI-keuzes onderbouwen en je team sneller laten ontwikkelen.
Andere vraag?
Niet elke opdracht past in een hokje. Vertel wat je wilt bouwen, dan kijken we wie uit ons netwerk erbij past.
Bespreek je vraag →Architectuur
Hoe goede Python-developers systemen ontwerpen
Drie patronen die we in bijna elke opdracht tegenkomen. Links de opbouw en de afwegingen, rechts de kern in Python, zoals je hem in een code review wilt zien.
Een API die drukte aankan
Een koppeling waar andere systemen op leunen, moet snel blijven onder piekbelasting en voorspelbaar falen.
- LatencyAsync I/O en connection pooling: één trage databasecall houdt de andere requests niet op.
- ValidatiePydantic controleert elk verzoek aan de rand. Ongeldige invoer krijgt direct een heldere 422.
- SchaalStateless workers achter een load balancer. Opschalen is replica’s bijzetten, geen herbouw.
from decimal import Decimal from fastapi import Depends, FastAPI from pydantic import BaseModel, ConfigDict, Field from sqlalchemy.ext.asyncio import AsyncSession app = FastAPI() class QuoteIn(BaseModel): customer_id: int amount: Decimal = Field(gt=0, decimal_places=2) class QuoteOut(QuoteIn): model_config = ConfigDict(from_attributes=True) id: int # validatie aan de rand: wat binnenkomt, klopt @app.post("/quotes", status_code=201) async def create_quote( data: QuoteIn, session: AsyncSession = Depends(get_session), ) -> QuoteOut: async with session.begin(): # één transactie per request quote = await quotes.add(session, data) return QuoteOut.model_validate(quote)
Data die je kunt vertrouwen
Een AI-toepassing is zo goed als de data eronder. Een pipeline moet fouten zichtbaar maken in plaats van ze door te geven.
- KwaliteitElk record wordt gevalideerd tegen een schema. Afwijkingen gaan naar quarantaine, niet ongemerkt verloren.
- HerhaalbaarUpserts op een vaste sleutel maken de run idempotent, dus opnieuw draaien na een storing is veilig.
- ObservabilityPer run gaan de aantallen geladen en geweigerd naar monitoring, zodat een daling direct opvalt.
from collections.abc import AsyncIterator from decimal import Decimal from typing import Literal from pydantic import BaseModel, ValidationError class Order(BaseModel): order_id: str total: Decimal currency: Literal["EUR"] async def load_orders(rows: AsyncIterator[dict]) -> RunStats: stats = RunStats() async for row in rows: try: order = Order.model_validate(row) except ValidationError as err: # afwijkingen apart zetten, nooit stil weggooien await quarantine.put(row, reason=str(err)) stats.rejected += 1 continue # idempotent: opnieuw draaien na een storing is veilig await warehouse.upsert("orders", order, key="order_id") stats.loaded += 1 return stats
Een agent met grenzen
Een agent die zelf acties uitvoert, heeft vaste kaders nodig: wat hij mag, tot welk bedrag, en wie meekijkt.
- KadersHet model kiest uit een vaste set acties en antwoordt in een getypeerd formaat. Vrije tekst voert niets uit.
- Mens beslistBoven een ingestelde grens gaat het besluit naar een medewerker. De agent bereidt voor, de mens keurt goed.
- ControleerbaarElke beslissing wordt gelogd met de gebruikte bronnen, zodat je achteraf kunt nagaan waarom.
from decimal import Decimal from typing import Literal from pydantic import BaseModel REFUND_LIMIT = Decimal("250.00") # het model kiest uit vaste acties, met een getypeerd antwoord class Decision(BaseModel): action: Literal["answer", "refund", "escalate"] amount: Decimal | None = None sources: list[str] async def handle(ticket: Ticket) -> Decision: context = await retriever.search(ticket.text, k=5) decision = await llm.parse( build_prompt(ticket, context), output=Decision ) if decision.amount and decision.amount > REFUND_LIMIT: return await escalate(ticket, decision) # boven de grens beslist een medewerker await audit_log.write(ticket.id, decision) # elke beslissing terug te vinden, met bronnen return decision
Waarom Python People
Gespecialiseerd in Python-developers
Vier redenen waarom opdrachtgevers via ons een Python-developer inhuren of werven.
Alleen Python
Van backend en data tot AI: we doen uitsluitend Python. Daardoor weten we precies wat een sterke developer onderscheidt.
Een eigen netwerk
We kennen de developers die we voorstellen, ook mensen die niet op vacatures reageren. Zo heb je binnen 48 uur een eerste selectie.
Technisch gescreend
Vakmensen toetsen elke kandidaat op wat die werkelijk heeft gebouwd. Jij hoeft de techniek niet zelf te beoordelen.
Vooraf duidelijk, flexibel opzegbaar
Je weet vooraf het all-in uurtarief, betaalt alleen gewerkte uren en zit niet vast aan een lange verbintenis.
Het model is zelden het probleem. Wat een AI-toepassing laat slagen, zit eromheen: data die klopt, een evaluatie die meet en beheer dat vanaf de eerste sprint meetelt.
Marktbeeld
Waarom zoveel AI-projecten blijven steken
Het model is zelden het probleem. De randvoorwaarden wel. Lees hoe je dat voorkomt
Veel pilots halen productie niet
Gartner voorspelde dat minstens 30% van de generatieve-AI-projecten na de proof of concept zou stoppen. De redenen: slechte datakwaliteit, onvoldoende risicobeheersing, oplopende kosten of onduidelijke waarde.
Bron: Gartner, 29 juli 2024 (opent in nieuw tabblad)Agents vragen om discipline
Volgens Gartner wordt meer dan 40% van de projecten met AI-agents eind 2027 stopgezet, om vergelijkbare redenen: kosten, onduidelijke waarde en onvoldoende risicobeheersing.
Bron: SD Times over Gartner, 30 juni 2025 (opent in nieuw tabblad)AI-geletterdheid is een plicht
Sinds 2 februari 2025 moeten organisaties die AI-systemen inzetten maatregelen nemen voor de AI-geletterdheid van hun medewerkers. Dat vraagt ook iets van je engineers.
Bron: AI-verordening, artikel 4 (opent in nieuw tabblad)Blog
Verdieping voor beslissers en developers
AI-agents bouwen in Python
Zes ontwerpregels voor agents die taken betrouwbaar afronden, zonder verrassingen in kosten of veiligheid.
LLM-toepassingen evalueren in Python
Waarom "het ziet er goed uit" niet genoeg is, en hoe je in vier stappen een evaluatie opzet die werkt.
AI-codingassistenten in je Python-team
Hoe je de snelheidswinst van codingassistenten pakt zonder in te leveren op kwaliteit en veiligheid.
Veelgestelde vragen
Inhuren en werven in het kort
Hoe huur ik een freelance Python-developer in?
Bel of mail ons; een vacaturetekst is genoeg. Binnen 48 uur krijg je een selectie van passende kandidaten. Je spreekt ze zelf, en kies je iemand, dan kan die vaak binnen een week beginnen.
Hoe snel kan een developer starten?
Vaak binnen een week nadat er een match is. Dat hangt af van de beschikbaarheid van de developer en van je eigen onboarding.
Wat kost een freelance Python-developer?
Dat hangt af van niveau en specialisatie. Met de calculator op deze pagina zie je direct een indicatief uurtarief, all-in en excl. btw. Je betaalt alleen de uren die gewerkt worden.
Hoe beoordelen jullie of een developer past?
We spreken elke developer zelf en toetsen op wat hij of zij werkelijk in productie bracht: hoe kwaliteit werd gemeten, hoe risico’s werden beheerst en wat er gebeurde toen het misging.
Werven jullie ook voor een vaste functie?
Ja. Via werving en selectie stellen we getoetste kandidaten uit ons eigen netwerk voor, voor een vaste plek in je team. 100% No Cure, No Pay, met een garantie in de eerste maand.
Contact
Op zoek naar een Python-developer?
Bel ons direct, of stuur je vraag. We nemen contact met je op.
Python · AI · data
Hire or recruit Python developers, specialised in AI
Python People has its own network of experienced Python developers. From it, we introduce freelancers for assignments and recruit developers for permanent roles in your team. Freelancers at senior and architect level, permanent roles from mid-level to architect. Specialised in AI, data and backend.
- A first selection within 48 hours, no obligation
- Flexible notice
- No cost for absence or holidays
A freelance Python developer for your assignment
Dutch-speaking freelancers from our own network of Python developers, interviewed and assessed by us. You receive a first selection within 48 hours, with no obligation, and you know the hourly rate upfront.
Best for: teams that need extra expertise or capacity for a while. Hire a freelancer Permanent recruitmentA Python developer on your payroll
From our own network of Python developers we introduce candidates for a permanent role in your team, including people who are not actively looking. Everyone is technically assessed. 100% No Cure, No Pay, with a guarantee during the first month.
Best for: organisations that want to keep the expertise in-house. Explore recruitmentAre you a Python developer? View assignments and jobs
In demand across our roles
- Python 3.12
- FastAPI
- Pydantic
- pytest
- PostgreSQL
- pgvector
- LangGraph
- Airflow
- dbt
- MLflow
- Kubernetes
- Azure OpenAI
Clients include
- Adyen
- Brink’s
- Eurofins
- Fox-IT
- GS1
- Insmed
- KPN
- Vacansoleil
Hire a freelancer
How it works
Four steps, from your first question to the first working day.
- 01
Tell us what you need
Call or email us. A job description is enough: we translate it into the right technical profile.
- 02
We introduce candidates
Within 48 hours you receive suitable candidates from our own network of Python developers.
- 03
Meet them, no obligation
You choose who to meet and hold the conversation yourself, so you quickly know whether someone fits your team.
- 04
The developer starts
Once you choose a candidate, they can often start within a week.
Permanent recruitment
Prefer someone permanent?
We take the recruitment process off your hands, from job description to first working day, with attention to your team and culture. 100% No Cure, No Pay.
- 01
Job description
You send us the vacancy or job description. In a short call we discuss your wishes, the team and the culture.
- 02
Our network
Candidates from our own network of Python developers, including people who are not actively looking.
- 03
Selection
We interview the candidates and assess their technical skills, so you only receive candidates who fully match.
- 04
Meet and start
You meet the best candidates. We guide the offer through to the first working day, with one fixed point of contact.
AI specialisms
Where our Python developers make the difference
An AI application touches your entire stack: from the data at the bottom to the API, the monitoring and the decisions on top. We select per discipline.
Agentic AI
Agents that complete tasks in your systems, with clear boundaries and full traceability.
RAG and knowledge systems
Reliable answers from your own documents, with citations and per-user permissions.
LLM integration
Language models as a dependable part of your existing software and processes.
AI backends with FastAPI
Fast, typed Python backends that serve your software and AI features securely and at scale.
Data engineering
Python pipelines that keep your data clean, current and securely available, as the foundation for every application.
MLOps and LLMOps
Evaluation, deployment and monitoring, so AI keeps performing after launch.
AI architecture
Senior architects who justify AI choices and help your team build faster.
Something else?
Not every assignment fits a box. Tell us what you want to build and we’ll find the profile to match.
Discuss your question →Architecture
How strong Python developers design systems
Three patterns we meet in almost every assignment. On the left the structure and the trade-offs, on the right the core in Python, the way you’d want to see it in a code review.
An API that handles the load
An integration other systems depend on has to stay fast under peak load and fail in predictable ways.
- LatencyAsync I/O and connection pooling: one slow database call doesn’t hold up the other requests.
- ValidationPydantic checks every request at the edge. Invalid input gets a clear 422 straight away.
- ScaleStateless workers behind a load balancer. Scaling up means adding replicas, not rebuilding.
from decimal import Decimal from fastapi import Depends, FastAPI from pydantic import BaseModel, ConfigDict, Field from sqlalchemy.ext.asyncio import AsyncSession app = FastAPI() class QuoteIn(BaseModel): customer_id: int amount: Decimal = Field(gt=0, decimal_places=2) class QuoteOut(QuoteIn): model_config = ConfigDict(from_attributes=True) id: int # validate at the edge: whatever gets in is correct @app.post("/quotes", status_code=201) async def create_quote( data: QuoteIn, session: AsyncSession = Depends(get_session), ) -> QuoteOut: async with session.begin(): # one transaction per request quote = await quotes.add(session, data) return QuoteOut.model_validate(quote)
Data you can trust
An AI application is only as good as the data underneath. A pipeline should surface errors instead of passing them on.
- QualityEvery record is validated against a schema. Deviations go to quarantine instead of getting lost unnoticed.
- RepeatableUpserts on a fixed key make each run idempotent, so re-running after a failure is safe.
- ObservabilityEach run reports loaded and rejected counts to monitoring, so a drop stands out immediately.
from collections.abc import AsyncIterator from decimal import Decimal from typing import Literal from pydantic import BaseModel, ValidationError class Order(BaseModel): order_id: str total: Decimal currency: Literal["EUR"] async def load_orders(rows: AsyncIterator[dict]) -> RunStats: stats = RunStats() async for row in rows: try: order = Order.model_validate(row) except ValidationError as err: # set deviations aside, never drop them silently await quarantine.put(row, reason=str(err)) stats.rejected += 1 continue # idempotent: re-running after a failure is safe await warehouse.upsert("orders", order, key="order_id") stats.loaded += 1 return stats
An agent with boundaries
An agent that takes actions on its own needs firm limits: what it may do, up to what amount, and who reviews.
- LimitsThe model picks from a fixed set of actions and answers in a typed format. Free text executes nothing.
- People decideAbove a set limit, the decision goes to a person. The agent prepares, the person approves.
- AuditableEvery decision is logged with the sources used, so you can trace afterwards why it happened.
from decimal import Decimal from typing import Literal from pydantic import BaseModel REFUND_LIMIT = Decimal("250.00") # the model picks from fixed actions, with a typed answer class Decision(BaseModel): action: Literal["answer", "refund", "escalate"] amount: Decimal | None = None sources: list[str] async def handle(ticket: Ticket) -> Decision: context = await retriever.search(ticket.text, k=5) decision = await llm.parse( build_prompt(ticket, context), output=Decision ) if decision.amount and decision.amount > REFUND_LIMIT: return await escalate(ticket, decision) # above the limit, a person decides await audit_log.write(ticket.id, decision) # every decision traceable, with its sources return decision
Why Python People
Specialists in Python developers
Four reasons why clients hire or recruit a Python developer through us.
Only Python
From backend and data to AI: we work exclusively with Python. That's why we know exactly what sets a strong developer apart.
Our own network
We know the developers we introduce, including people who don't respond to job ads. That's how you get a first selection within 48 hours.
Technically screened
Specialists assess every candidate on what they have actually built. You don't have to judge the technology yourself.
Clear upfront, flexible notice
You know the all-in hourly rate upfront, only pay for hours worked and aren't tied to a long commitment.
The model is rarely the problem. What makes an AI application succeed is everything around it: data you can trust, an evaluation that measures, and operations that count from the first sprint.
Market view
Why so many AI projects stall
The model is rarely the problem. Everything around it is. Read how to avoid it
Many pilots never reach production
Gartner predicted that at least 30% of generative AI projects would be abandoned after proof of concept, citing poor data quality, inadequate risk controls, escalating costs or unclear business value.
Source: Gartner, 29 July 2024 (opens in a new tab)Agents demand discipline
According to Gartner, over 40% of agentic AI projects will be cancelled by the end of 2027, for similar reasons: escalating costs, unclear business value and inadequate risk controls.
Source: SD Times on Gartner, 30 June 2025 (opens in a new tab)AI literacy is now a legal duty
Since 2 February 2025, organisations deploying AI systems in the EU must take measures to ensure the AI literacy of their staff. That applies to your engineers too.
Source: EU AI Act, Article 4 (opens in a new tab)Blog
In-depth reading for decision-makers and developers
Building AI agents in Python
Six design rules for agents that complete tasks reliably, without surprises in cost or security.
Evaluating LLM applications in Python
Why "it looks good" isn't enough, and how to set up an evaluation that works in four steps.
AI coding assistants in your Python team
How to capture the speed gains of coding assistants without compromising on quality and security.
Frequently asked questions
Hiring and recruiting in brief
How do I hire a freelance Python developer?
Call or email us with the technologies, the project, the duration, the level and the work location. Within 48 hours you receive a selection of suitable candidates. You talk to them yourself, and once you choose someone, they can often start within a week.
How quickly can a developer start?
Often within a week of finding a match. It depends on the developer’s availability and on your own onboarding.
What does a freelance Python developer cost?
It depends on level and specialism. The calculator on this page gives you an indicative hourly rate straight away, all-in and excl. VAT. You only pay for the hours worked.
How do you judge whether a developer fits?
We talk to every developer ourselves and assess what they actually took to production: how quality was measured, how risks were managed and what happened when things went wrong.
Do you also recruit for permanent roles?
Yes. Through permanent recruitment we introduce assessed candidates from our own network for a permanent place in your team. 100% No Cure, No Pay, with a guarantee during the first month.
Contact
Looking for a Python developer?
Call us directly, or send your question. We’ll get in touch with you.
Cookies
Cookie Policy pythonpeople.nl
Last updated: 22 November 2026
1. Who we are
pythonpeople.nl is a website of IT Recruitment BV, trading as Python People. We are responsible for the use of cookies and similar technologies on our site.
- Address: Herenweg 115, 2105 MG Heemstede, the Netherlands
- Chamber of Commerce (KvK) number: 60485965
- VAT number: NL853931501B01
- E-mail: info@pythonpeople.nl
This cookie policy applies to pythonpeople.nl and to everyone who visits or uses our site: visitors, freelancers and other candidates, and clients. It accompanies our Privacy Statement, in particular the section ‘Cookies en analytics’ (Cookies and analytics).
2. What are cookies and similar technologies?
Cookies are small text files that your browser stores on your computer, tablet or phone when you visit a website. In addition to cookies, we may use local browser storage (localStorage), for example to remember your cookie choice. Some services use similar technologies, such as pixels or tags. In this policy, ‘cookies’ also includes those similar technologies.
The retention period in the table below is how long the file stays on your device. How long we or our service providers keep the data collected is explained in the privacy statement under ‘Hoe lang we gegevens bewaren’ (How long we keep data).
3. Which cookies do we use?
We use two types of cookies:
- Strictly necessary cookies. These are needed for our site to work, to keep it secure, and to remember your cookie choice. These cookies do not require consent (section 11.7a of the Dutch Telecommunications Act).
- Statistics cookies. These help us understand how our site is used, so that we can improve it. We place these cookies only if you give your consent (section 11.7a of the Telecommunications Act and Articles 6(1)(a) and 7 GDPR).
We do not currently place marketing or advertising cookies. If we decide to do so later, we will update this policy first and ask for your separate consent. Consent for statistics will then not count as consent for marketing.
4. Your choice: give, refuse or change consent
When you first visit, you will see a cookie banner. In that banner you can:
- Accept all: you consent to statistics cookies.
- Reject all: we place only strictly necessary cookies. You can still use our site, including the forms.
Rejecting is just as easy as accepting. The buttons are equivalent: the same size and equally prominent. Until you make a choice, we do not place statistics cookies and we do not collect statistics. Scrolling or clicking further is not consent. There are no pre-ticked options.
Accepting our general terms and conditions, submitting a form, uploading your CV or using our tools or checks does not count as consent to statistics cookies.
Changed your mind? You can change or withdraw your choice at any time via Cookie settings at the bottom of every page. If you withdraw your consent, statistics stop immediately and no new data are collected. Withdrawal applies going forward and does not affect processing that already took place with your consent. You can also delete cookies already placed via your browser settings (see section 10).
Your choice applies per browser and per device. If you clear your cookies or use another device, we will ask for your choice again.
5. Cookie overview
The table below is the overview we will use at launch. We will verify the technical names and exact lifetimes immediately after launch and update this table if needed.
| Cookie | Type | Placed by | Purpose | Lifetime |
|---|---|---|---|---|
| Cookie preference | Strictly necessary | pythonpeople.nl | Remember the choice you made in the cookie banner, so that the banner does not keep reappearing | 12 months |
| Cookie preference (local storage) | Strictly necessary | pythonpeople.nl | A copy of your cookie choice in your browser’s local storage | For as long as your choice remains valid |
| Statistics session cookie | Statistics, only after consent | External analytics service, via pythonpeople.nl | Link page views within a single visit | Typically 30 minutes, to be confirmed after launch |
| Analytics visitor cookies | Statistics, only after consent | External analytics service, via pythonpeople.nl | Distinguish returning visitors and visits for statistics | Typically up to 14 months, to be confirmed after launch |
You can look up the technical cookie names in your browser settings. On request we will also send you an overview; e-mail info@pythonpeople.nl.
Again: the periods in this table are the lifetimes of the cookies on your device, not the retention periods of the underlying data. Those are explained in the privacy statement under ‘Hoe lang we gegevens bewaren’ (How long we keep data).
6. Statistics: what we measure and how
If you consent to statistics cookies, we measure, among other things:
- which pages, vacancies and assignments are viewed;
- how visitors reach our site (for example via a search engine or a link);
- device type and browser;
- general actions on our site, such as opening a form or viewing the blog.
For this we use an external analytics service, which our technical service providers configure or manage for us. We configure it as privacy-friendly as possible:
- statistics are loaded only after you have given consent;
- your IP address is not stored, and is used only briefly to determine an approximate location (country or region);
- we do not send names, e-mail addresses, CVs or other directly identifying data to the analytics service;
- we use statistics data only to improve our site, not for advertising.
We will confirm the exact settings at launch (including IP-address handling and how long the analytics service keeps the data) and align them with the privacy statement.
7. Forms, CVs and additional features
On our site you can send us a message, submit a search request or rate enquiry, apply for a role and upload your CV. If we offer tools or checks, for example a rate-indication helper or another check, you can use those there as well.
Those features process the data you enter yourself. How we handle that is explained in the Privacy Statement. Using these features does not in itself place statistics cookies and does not count as consent for them. If you have refused statistics cookies, the forms and any tools will still work.
If a tool or check later loads an additional script or a third party, we will mention that first in this policy and in the privacy statement and, where required, ask for separate consent.
8. Service providers and data outside the European Economic Area
For the technology, hosting and statistics of our site we work with technical service providers. With the providers that process data on our instructions, we have arrangements on the protection of your data, including a data processing agreement. They may use the data only on our instructions. You can read more in the privacy statement under ‘Met wie we gegevens delen’ (Who we share data with).
Some of our service providers may process data outside the European Economic Area (EEA), for example in the United States. That happens only with appropriate safeguards: an adequacy decision of the European Commission (such as the EU-US Data Privacy Framework, only for recipients certified under it) or the European Commission’s standard contractual clauses, with supplementary measures where needed. You can read more in the privacy statement under ‘Gegevens buiten de Europese Economische Ruimte’ (Data outside the European Economic Area).
9. External websites and third-party content
Links to other websites. Our site, for example the blog or a vacancy or assignment, may contain links to other websites, such as a client’s site or an external application system. If you click those links, you leave pythonpeople.nl. That website’s own cookie and privacy policies apply. We have no control over them. An ordinary link does not place cookies until you click it.
Content on our site. We publish the vacancies, assignments and profiles on our site ourselves. If a client or user supplies content to us, or if you can later post content yourself (such as a comment), that content must not contain scripts, tracking pixels, tags, hidden iframes or other technologies that place cookies or collect data about visitors. Ordinary links are allowed. We may edit or remove content that does so anyway.
Accounts. You cannot create an account or log in on our site. If we offer this later, we will update this policy first.
10. Managing cookies in your browser
In your browser settings you can view, block and delete cookies and local storage. Please note:
- If you block strictly necessary cookies, features such as remembering your cookie choice or the forms may not work properly.
- If you delete only the cookies, your previous choice may disappear and we will ask for consent again. To make sure that statistics are no longer collected, also update your choice via Cookie settings.
You can find guidance in your browser’s help pages, for example by searching for ‘delete cookies’ and the name of your browser.
11. Your privacy rights
Statistics cookies may involve the processing of personal data. You then have the rights set out in the GDPR, such as the right of access, erasure and objection, and the right to withdraw your consent. You may also lodge a complaint with the Dutch Data Protection Authority (Autoriteit Persoonsgegevens). How to exercise these rights is explained in the Privacy Statement under ‘Je privacyrechten’ (Your privacy rights).
12. Changes and contact
This cookie policy is available at pythonpeople.nl/cookiebeleid. We will update it if we start using other cookies or if the law changes. The date at the top shows when the policy was last updated. If we want to use cookies for a new purpose that requires consent, we will ask for your consent again first.
Questions about this cookie policy? E-mail info@pythonpeople.nl or write to IT Recruitment BV, trading as Python People, Herenweg 115, 2105 MG Heemstede, the Netherlands.