Machine learning companies
HR policies and employee handbook
The handbook for a machine learning firm needs sections that other firms do not. We write the data-use and ethics conduct policy, including how a concern is raised and answered. We add the compute policy, the annotator levels and pay review rule, and the model upkeep ownership rule. We include confidentiality terms for staff who handle client data, which your own adviser confirms. In the HR system we record acknowledgements, review dates and the annotator training record, so nothing depends on memory. Each policy is short enough to read in one sitting.

Find the service that fits your priority.
Which rule is being interpreted differently by employees, supervisors or locations? Each option below has a separate scope; you do not need to buy the full group.
HR Policies
GullyHR drafts and reviews written HR policies for growing companies. We agree which of thirty policies you need first, settle the choices management must make, and draft each in one structure with a procedure, forms and an employee FAQ. Your own legal adviser reviews every draft. This page is the hub for all thirty policy guides.
Scope, approach and deliverables →
HR Policy Packages
HR policy packages are defined sets of written policies that GullyHR drafts for a growing company: Starter has five, Core has eleven and Complete has all thirty. Every policy follows one section structure with an employee FAQ, a workflow and forms. The drafts go to your own legal adviser before issue, and the fee is agreed after a short call.
Scope, approach and deliverables →
Employee Handbook
An employee handbook is one plain-language document that brings a company's key policies, practices and contacts together for its employees. GullyHR drafts it for growing companies from their written policies, with a welcome section, short policy summaries, a contacts list and an acknowledgement page. Your own legal adviser reviews it before issue.
Scope, approach and deliverables →
Ground the work in your operating needs.
A machine learning company builds models and systems that learn from data. Here is how to manage data scientists, ML engineers and annotators where research is uncertain, compute is costly and good people are very scarce.
Prepare for the discussion
- Current policies, appointment templates and employee communications
- Examples of unclear rules or inconsistent decisions
- Your legal or compliance review contact and the person authorised to approve policies
Policies must be reviewed for your establishment, jurisdiction, contracts and workforce. A standard package is a starting structure, not legal advice or automatic confirmation of compliance.
Before you choose
Start with the issue that is interrupting work or leaving a decision unclear. Which rule is being interpreted differently by employees, supervisors or locations? Ask the team to name one failed experiment from this quarter and what it taught them.
Start with the decisions causing confusion and any essential gaps identified with your advisers. A handbook becomes useful when it brings those approved policies into one coherent employee reference.
Yes. The 3 options are a menu, not a requirement to buy a package. We agree the problem, deliverables, responsibilities and exclusions before work begins.
Tell us that you operate in machine learning companies, your team size and locations, the service you are considering and one example of the issue. Do not attach personal employee records to the initial enquiry; any later evidence sharing needs agreed access and handling.
They follow the locations, workforce, available records and deliverables in scope. The first conversation establishes those details; a written scope and fee are agreed before paid work starts.
Start with the issue you need to resolve.
Tell us about your machine learning companies business, the teams involved and the help you need. We’ll use that to agree the right scope.