Github software uk ltd companies house
Come build the home for all developers
Do the best work of your career and join in our mission to accelerate human progress by connecting communities all over the world through software collaboration.
Open positions Internships
A diverse and inclusive workplace
At GitHub, we think that a diverse company is a strong company, and we work hard to foster a supportive and welcoming workplace. Learn more about our commitment to diversity.
Build amazing things with a balance of autonomy and collaborative teamwork. Set your own work schedule and make use of a flexible PTO plan when you need to recharge.
Lead from any location
GitHub is a remote-first company with offices located throughout the US, Europe, and Asia. Whether you live near an office or not, GitHub believes you can do your best work wherever you are. If you work remotely, you will receive a stipend to outfit your home office and receive reoccurring reimbursement refreshes.
Put your health and family first
Employees in the United States enjoy 100% coverage of health insurance premiums across our medical, dental, and vision plan offerings, including coverage for dependents. We also offer five months of paid family leave to all new parents with the option to use it all at once or throughout the baby’s first year.
Find your zen
GitHub provides a monthly wellness stipend designed to cover anything from gym memberships, massage, meditation apps, or any other wellness related expenses.
Invest in your future
At GitHub, you’ll have a stake in the future success of our platform with equity grants. For full-time employees in the United States, we offer competitive 401k planning with a 50% company match up to the IRS 402(g) annual limit.
Learn how you learn best. From books to conferences, you’ll get a yearly budget for your individual learning and development goals.
Give back to your community
We believe in sharing our time, resources, and products to contribute to positive social impact. GitHub matches charitable donations up to $15,000 per calendar year. And for each hour (up to 40 hours) of volunteering per year, you will receive $20 to donate to an organization of your choice.
Where we can hire
We love our remote culture at GitHub and we currently have hiring entities in all the following states and countries.
We hire in all 50 states
- British Columbia
Please note that benefits vary by country, the ones described on this page are for our US based employees. Benefit information for non-US based positions will be provided to individuals who interview for those roles.
No coffee runs here! We love watching talented people learn and explore their skills. Our interns gain real work experience by pairing up with GitHubbers on real projects. Check out what past interns have worked on.
Applications are currently closed for internships. Sign up to receive updates about internship opportunities via email.
A wrapper for the UK Companies House API, written in Python with tools for network analysis built on networkx.
This tool enables users to build a corporate network from information on the UK Companies House API.Networks are constructed through the following process:
- The user inputs a valid UK company number, alongside an API key and a number of desired iterations.
- From this company, chpy draws down information on its officers and persons of significant control (PSC).
- Relationships between officers/PSC and the company being searched are added to an edge list.
- If an officer is a valid UK company, it is added to a list of companies for examination in later iterations. NB: In 0.1.1, this functionality does not extend to PSCs.
- chpy then searches the Companies House API for each officer/PSCs appointments to other companies, using fuzzy matching and date of birth/address checks to verify search results.
- These relationships are then added to the edge list, and the companies added to the company table for analysis in later iterations.
- The process loops to a depth specified by the user.
New in 0.1.1
- The main function — get_company_network() — has been completely overhauled with:
- A new, clearer structure for data acquisition.
- Iteration depth.
- Output to networkx graph and pandas dataframes from within Python.
- Clearer output whilst running, however progress bar functionality has been suspended for the time being.
How it works
This tool is currently intended for use alongside a Jupyter notebook, and I’ve provided a sample in the “example” directory.I believe that most users will want to do with this is simply build a network from a company that they have interest in.It is strongly recommended that the depth is set to either 1 or 2. Depth scales exponentially, as will errors (see “A word of warning”, below).
pip install chpy
from chpy.build import *# You'll need your own API key from Companies House and to load# it as a variable in Python.with open('API_KEY.txt') as f: api_key = f.read().strip()graph, edge_list, company_table = get_company_network("a valid company number", api_key, 2)
The above code returns a graph in networkx format, and an edge_list and company_table as Pandas dataframes.
Additionally, chpy outputs three objects to ./data/company_number_depth/:
- One node list in csv format
- One edge list in csv format
- One graph in gexf format, for use with gephy or similar
A word of warning
This is very much still in development, and due to the way Companies House (CH) data is maintained and structured a degree of caution is required when using this tool. Notably:
- CH does not maintain information on companies that have been wound up for a certain period of time, so be aware that in many cases the data produced will be incomplete.
- Similarly, inactive officers are not always available.
- This software makes extensive use of fuzzy matching to build links between companies and officers, due to the limitations of the unique identifier system used by the CH API. The system is designed to be conservative with the relationships it builds, requiring date of birth (if available) and location matches, and as such may miss relationships if a strong match isn’t found.
- Similarly, given the use of the CH search API and fuzzy matching, despite the safeguards outlined above, “false positives” may still occur in exceptional circumstances.