Skip to content
deciml

Privacy Policy

Last updated: November 2026

Deciml ("we", "us", "our") operates the Deciml recruitment platform at deciml.io. This Privacy Policy explains how we collect, use, store and protect your personal data when you use our platform, in accordance with the UK General Data Protection Regulation (UK GDPR), the EU General Data Protection Regulation (EU GDPR) where applicable, and the Data Protection Act 2018.

1. Data Controller

Birch & James Associates Limited, registered in England and Wales (company no. 09549374), registered office: Office Gold Building 7, Floor 5, 566 Chiswick High Road, Chiswick Business Park, London W4 5YG, United Kingdom, trading as deciml and TechChain Talent, is the controller for the processing described in this notice where we operate our own candidate profiles and recruitment services, manage company accounts and business contacts, or determine other purposes identified below. Where you apply directly to a company's role through its company applicant-tracking workspace, that company is normally the controller for your application and we process it on that company's instructions.

We remain the controller of our own agency recruitment records, including our recruiters' notes and the AI-assisted assessments we share with a hiring client when we introduce you (see Sections 4 and 5). If you have questions about this policy or wish to exercise your rights, please contact us at privacy@deciml.io.

Our representative in the European Union, under Article 27 of the EU GDPR, is Orsa Saiwai EDS Limited (registered in Ireland, company no. 514924), 2 Seapoint Avenue, Blackrock, Dublin, A94 VY68, Ireland. If you are in the European Union you may contact our representative, in addition to or instead of us, about any question on how we process your personal data. You can also write to us at privacy@deciml.io and we will involve them.

A company using deciml's applicant tracking system may add candidates it is recruiting itself. For those records the company is the controller and Birch & James Associates Limited (deciml) is its processor. deciml does not email those candidates, add them to its candidate pool or match them to other roles. To exercise your rights over such a record, contact the company that added you; you can also write to us and we will pass your request on. The company also decides how long we keep those records: between 6 months and 2 years after the last recruitment activity on an application (such as a stage change, note, interview or message), and until its agreement with us ends for someone it hired. When the period runs out we delete the application, and with a person's last application their record and CV. When a company's agreement with us ends, it can download those records for 30 days, and then we delete them all.

Applications to company-controlled roles

If you apply directly to a company's role using its deciml-powered careers page or applicant-tracking workspace, the company named on the role is responsible for deciding how your application is used and retained.

We provide the technology and process your application on the company's instructions. Its authorised users can review your application. Our own agency recruiters do not have routine recruitment access to direct applicants in STANDARD company workspaces. Limited authorised technical or security processing may be necessary to operate and protect the service.

Contact the company identified on the application page about access, correction, deletion or other rights relating to its application record. If you contact us, we will pass the request to the company and assist it.

2. What personal data we collect

2.1 Candidate data

When you create a profile on our platform, we collect:

  • Name, email address and password (hashed)
  • Professional information: job title, location, bio, work experience, education
  • Skills and proficiency levels
  • Career preferences: desired role, work type (remote/hybrid/onsite), salary expectations, preferred location and company stage
  • Availability status (actively looking, open to opportunities, not looking)
  • Social and messaging handles: Telegram, Discord, GitHub, Twitter/X
  • If you verify a GitHub or X handle (Run my match): the handle, when it was verified, and for X, the links to up to five public posts you chose to submit and their text
  • Scores we calculate from your public GitHub activity, and topic labels for the X posts you submitted. The topic labels come from an AI model that sees only the text of each post, never your name or handle; they describe your public work and are not used to accept or reject you for any role
  • CV/resume documents you upload
  • Profile photograph
  • Application history and cover notes
  • Communication preferences (email and Telegram notification settings)
  • If you answer it when you sign up: how you heard about deciml, and any words you add to "Other". It is optional, and we use it only to understand which channels bring candidates to us

2.2 Data from CVs

When you upload a CV, we use AI-powered text extraction to parse your document and pre-populate your profile with information including your name, title, skills, work history and education. You can review and edit all extracted data before it is saved.

2.3 Technical and usage data

  • IP address, browser type and version
  • Pages visited, time spent on pages, referral source
  • Device information (operating system, screen resolution)
  • If you are a candidate: the days you said GM (checked in), to show your streak to you and to recruiters

Usage analytics are collected first-party and processed by PostHog on EU-hosted infrastructure acting as our processor. We do not use advertising pixels, cross-site trackers, or browser-extension telemetry, we do not run third-party ad networks on any page, and analytics data is never sold or shared with advertisers.

2.4 Data from third-party integrations

  • Telegram chat ID (when you link your Telegram account for notifications)
  • Job data synced from our CRM system (Manatal) for the roles listed on the platform

2.5 People we contact about hiring

Our recruitment business, trading as TechChain Talent, contacts founders, executives and hiring managers at companies that are likely to be hiring, usually after a funding round, to offer our recruitment services. If you received an email or message from us about hiring, this section explains what we hold about you and why.

  • What we hold: your name, job title and seniority; your employer, its funding rounds and investors; your work email address and whether it has been verified; where available, your phone number, location, education, LinkedIn, X and Telegram profiles, your X follower count and an influence score we calculate from it; notes our team writes about you; the messages we draft or send to you, including passages drafted with the help of AI, and your replies; and labels we use to prioritise our outreach.
  • Where it comes from: RootData (public funding data, including founders and team members and their public profiles), Apollo (a business-contact data provider), public web sources found by our AI-assisted research tool, lists our team imports or enters by hand, and anything you send us, such as an enquiry through our website. We aim to email work addresses only, and hold back any address at a known personal email provider.
  • Why, and our legal basis: to offer recruitment services to your company. Our legal basis is legitimate interests: approaching people in their professional capacity about hiring for the company they work for.
  • How we contact you: a person on our team decides to contact you, and nobody is added to our outreach without that decision. Messages on LinkedIn, Telegram or by phone are always sent by a person; after that decision, some emails in a sequence may be sent on a schedule. Parts of our emails are drafted with the help of an AI model, which sees your name, job title and facts about your company; your email address and contact handles are not shared with it.
  • Who we share it with: Apollo (contact data), Google Workspace (the mailbox our emails are sent from and replies arrive in), OpenRouter/Anthropic (drafting and research), Slack (internal alerts to our team about new funding rounds, which can include your name and contact details), and the providers that run our platform: Supabase (database), Vercel (hosting), Inngest (background jobs) and Sentry (error monitoring). We never share it with other clients or candidates.
  • How long we keep it: we delete your details once 24 months have passed without us contacting you or recording anything new about you, such as a new funding round at your company. If you reply to us or your company becomes a client, we keep them while we work together. If you opt out, we keep only what we need to make sure we never contact you again. If you ask us to erase your details, we keep only your email address and a coded form of your LinkedIn profile address (whichever of the two we held), so that we never add or contact you again. Emails we exchanged also remain in our email mailbox.
  • Opting out: use the unsubscribe link in any of our emails, reply asking us to stop, or write to privacy@deciml.io. We then stop all contact on every channel. Your other rights are set out in Section 8.

2.6 Card verification and scans

Card verification is off unless you switch it on. When it is on, the QR code on your Deciml card opens a page that shows your card name (never your real name), finish, rarity, role family, a broad region, your availability, the signals we verified (GitHub, wallet, on-chain score, X posts, dated experience) and your card stats. It never shows your contact details, handles, wallet address, employer or exact location. The page carries a signature so a viewer can check the numbers have not been altered.

When someone opens that page, we may record a scan: the date, the city and country our hosting provider derives from their connection, and whether they came from the QR code, a link or a banner. We never store their IP address. Only you see your scans; recruiters never do. If the visitor creates an account through the sign-up link on that page within 7 days of the scan, we note that on the scan and, if your profile is public, credit you as their referrer. That link carries an opaque scan reference in its address; we set no cookie for it. If the visitor also arrived through a member's share link, that member is credited instead (a share link takes precedence over a card scan); the scan is still noted.

A visitor can ask us for an introduction from that page by leaving their email address and, optionally, a company and a note. We pass the request to our team and tell you someone asked; we ask you before we share any of your details. We keep the visitor's request as a notification to our team.

2.7 Anonymised profiles shown to companies

Anonymised, non-identifying summaries of profiles may be shown to companies considering hiring through us. A summary carries only a role family, a broad region, a band of years of experience, five skills and one line of evidence (for example "GitHub verified"), and each summary is shown only when at least five candidates in our network share it, so it cannot single anyone out. It never includes a name, photo, employer, school, handle, wallet, exact location, salary, notice period or availability. Imported profiles and candidates who have blocked any company are never included, and neither are hidden or closed profiles or anyone not looking for work. Nobody is introduced to a company without our asking them first.

When a company pastes a job description into our analyser, we read the role's facts with an AI model (company facts only: no candidate data goes into that step) and keep a record of what was parsed, never the job description itself. If the company books a call, we keep the parsed summary with their enquiry.

2.8 If you email one of our recruiters

Our recruiters can choose to have new emails in their own work mailbox sorted into a ranked list of tasks, so that they reply to people sooner. If you email a recruiter who has switched this on, an AI model (Anthropic, reached through OpenRouter, in the United States) is shown your display name, the subject line and the short preview line of your email. It is never shown your email address or the rest of your message. We keep only the resulting task: a one-line summary that does not quote your email, your name or company as a label, and a link to the email in the recruiter's mailbox. Finished tasks are deleted after 30 days. Open tasks are dismissed after 90 days and deleted 30 days later. All of them are deleted when the recruiter switches the feature off. Our legal basis is legitimate interests: you chose to write to a named recruiter about work, and this only helps them answer you. No decision about you is made by it. You can object, or ask what we hold, at the address in Section 8.

2.9 If you sit on a client's interview panel

A company that hires through our platform can put people on an interview panel, including people without a deciml account, by name and work email address. We hold that name and address, the panel role, the interview it is for, and the scorecard you submit, so that the hiring team can see who is on the panel and collect each person's scorecard, and so that we can email you the interview brief an hour before, a link to write your scorecard without an account, and up to two reminders if it is late. The company that added you decides who is on its panels; our legal basis is legitimate interests, supporting the hiring process you take part in for that company. We keep these details for as long as the company keeps the interview. You can ask what we hold, or ask us to erase your name and address, at the address in Section 8.

2.10 Applications to company-controlled roles

If you apply directly to a company's role using its deciml-powered careers page or applicant-tracking workspace, the company named on the role is responsible for deciding how your application is used and retained.

We provide the technology and process your application on the company's instructions. Its authorised users can review your application. Our own agency recruiters do not have routine recruitment access to direct applicants in STANDARD company workspaces. Limited authorised technical or security processing may be necessary to operate and protect the service.

Contact the company identified on the application page about access, correction, deletion or other rights relating to its application record. If you contact us, we will pass the request to the company and assist it.

2.11 Company account holders and administrators

Birch & James Associates Ltd is controller for the account information we use to create, assess, approve, administer and secure company access to the platform and communicate about the service. At sign-up we collect your name, work email, password and company name and website. After that we also hold your role in your company's workspace, your sign-in and security settings (such as two-factor authentication), the date of your last sign-in, and a record of the actions you take in the workspace with the IP address they came from, which our team uses to review sign-ups, support your company and keep the platform secure. The providers that run our platform (Section 5) process this information for us.

Our legal basis is legitimate interests: providing, securing and communicating about the service your company uses. If we decide not to approve a company that signed itself up, we delete its account 90 days after that decision, or 90 days after sign-up if it is still not approved, as described in Section 7; a sign-up whose email address is never verified is deleted after 7 days. The record of actions is deleted after 365 days. We keep an approved company's account information while the account is open and for up to 12 months after a seat is removed or the company's agreement with us ends; contract and invoice records are kept for six years. Deleting a removed seat's account after 12 months is not yet automatic: until it is, we delete it when the person or their company asks us at privacy@deciml.io. Your company's applicant retention setting does not determine how long we retain account-administration records. Your rights, and how to complain, are set out in Section 8.

When a company signs up, our team is notified with the company name and email domain so it can review the sign-up, and if we decide not to approve it we email the person who signed up to say so. While sign-up is by request, the name, email address and company details you send on the request form go to our team so we can reply, and we keep a record of the request (company name, website, email domain and the IP address it came from) for 365 days.

When an administrator accepts our Client ATS Terms for a company, we keep a record of the acceptance: the administrator's full name, work email address, IP address and browser details, the company's registered legal name and number, the versions accepted and the time. We keep it for the life of the company's agreement with us plus six years, as evidence of what was agreed.

3. How we use your data

We process your personal data for the following purposes:

  • Account creation and authentication: to create and manage your account on the platform (legal basis: contract performance)
  • Profile and job matching: to match your profile to relevant job opportunities using our automated matching engine, and to display your profile to our recruitment team (legal basis: contract performance and legitimate interests)
  • Application processing: to process your job applications and manage your candidacy through the recruitment pipeline (legal basis: contract performance)
  • Communications: to send you notifications about application status updates, new job matches, and weekly job digests via email and/or Telegram (legal basis: consent and legitimate interests)
  • Platform improvement: to analyse usage patterns and improve our platform and matching algorithms (legal basis: legitimate interests)
  • Security: to protect the platform against fraud, abuse and security incidents (legal basis: legitimate interests)
  • Legal compliance: to comply with applicable laws and regulations (legal basis: legal obligation)

4. AI-assisted recruitment

We use AI to compare candidate information with a role's requirements and produce a must-have checklist and written strengths and gaps. These assessments support recruitment review. Where we introduce you to a hiring client, we may also show that client the role-specific checklist and strengths and gaps alongside your profile and CV. Clients are not shown a numerical suitability score. AI outputs may be inaccurate or incomplete; contact privacy@deciml.io if you wish to correct information or raise a concern about an assessment.

Our platform uses automated processing to match candidate profiles to job opportunities and to score the quality of matches. This processing uses a combination of rules-based classification (matching your skills, experience and preferences against job requirements) and AI-powered scoring (using large language models to assess match quality). The AI model sees your job title, bio, skills, work history and education (the degree and subject only, not where or when you studied), and, if you have connected them, the scores we compute from your public GitHub and on-chain activity with a few short highlights (for example how many repositories you have or which Web3 organisations you have contributed to). It never sees your name, contact details, CV file, wallet address or social handles. It is not given your location as a field: whether you can work where a role is based is checked by fixed rules, not by the AI model. A place you mention in your title, bio or work history is sent as you wrote it. Before it reads your title, bio and work history, we remove your name (unless it is also part of a company, skill or job title you list), words that state your gender, and birth or graduation years, wherever our checks recognise them in what you have written. The company names and skills you list are sent as you wrote them.

These automated matches are used to surface relevant opportunities to you and to assist our recruitment team in identifying suitable candidates. No automated decision produces legal effects or similarly significant effects on you without human review. All hiring decisions involve human assessment by our recruitment team and the hiring client.

Our legal basis for this processing is the one given for profile and job matching in Section 3 (contract performance and legitimate interests). The AI model is reached through OpenRouter (Anthropic), and text from your profile is sent to OpenAI to create the embeddings matching uses; both are service providers listed in Section 5.

You have the right to request human review of any automated matching decision, to express your point of view, and to contest the outcome. Contact us at the email address above to exercise this right.

For accountability, we keep a log of automated matching decisions that were surfaced or acted on (the job, the score, and the model and version that produced it). This log is included in your data export (Section 8), is pseudonymised if you delete your account, and is retained for 24 months.

5. Who we share your data with

  • Hiring clients: For our recruitment introduction services, we share candidate information with the relevant hiring client under the introduction arrangements and permissions explained in this notice. This can include your name, title, skills, experience, location, CV, our recruiters' notes made visible to that client under the applicable feedback-visibility settings, and a role-specific AI-generated must-have checklist with written strengths and gaps. The client is not shown a numerical suitability score. The client uses the information for its own hiring assessment and is responsible for that use as controller.This differs from an application you send directly to a company's own role through its ATS workspace. For that application the company controls its recruitment record and we provide processing services on its instructions, as explained in the company-controlled applications section (Section 2.10).We do not share your salary expectations or availability status with clients without your consent.
  • Screening questions: When you apply for a role a company posted on deciml, the company may ask screening questions. Your answers, and whether an answer matched a requirement the company set, are shared with that company. We do not use your answers in our AI matching.
  • Duplicate submissions: If our recruiters put you forward to a company that already has you from its own channels, or the company later receives you itself, we record that both happened, on which dates, and who was first, matched on your email address only. We tell both our recruiters and the company that the other also has you; neither record is removed. We keep this record with your application to that company and delete it with it.
  • Public profile (optional): If you switch on a public profile, the profile sections you choose to make public (such as your name, headline, skills, credentials and scores) are visible to anyone at your chosen deciml.io/u/ address. Public profiles are off by default, your availability, salary expectations and applications are never public regardless of your settings, and you can switch the public profile off at any time. The page is shown only to you, never to anyone else, while your profile is closed or while we have hidden it (for example as a suspected fake account). If two of your profiles are merged, nobody else can take the address of the one merged away, and if that profile was public its address keeps pointing at your merged profile.
  • Service providers: We use the following third-party processors to operate the platform:
    • Supabase (database hosting and authentication, EU/US)
    • Vercel (application hosting, global CDN)
    • OpenRouter/Anthropic (AI-powered matching and CV parsing, and sorting emails sent to our recruiters, see Section 2.8)
    • OpenAI (text embeddings for matching and search, in the United States: we send the text of your profile's title, skills, work history, bio and, for short profiles, education, and the text of job descriptions and of recruiters' searches, so that similar roles and candidates can be found; we never send your name, contact details or CV file)
    • Resend (transactional email delivery)
    • Telegram (messaging notifications, where you have opted in)
    • Upstash (rate limiting infrastructure)
    • PostHog (first-party product analytics, EU-hosted)
    • Cloudflare (Turnstile bot protection on public forms, and R2 storage for encrypted backups of our database and stored files, including CVs, in Western Europe)
    • Sentry (error monitoring, EU-hosted; it also receives the security reports your browser sends when a page blocks a resource, which carry the page address, your browser and, by connecting, your IP address)
    • Inngest (background job processing, such as matching and notifications)
    • Google Workspace (our team's email and calendars, including interview invitations that carry your name and email address)
    • GitHub (our source code and automated testing; our nightly backup job copies our database and stored files through its servers before encrypting them, and our backup restore checks run there too, so a copy passes through them briefly)
    The sub-processors we use for a company's own ATS workspace, with each one's contracting entity, location and transfer mechanism, are listed at deciml.io/legal/sub-processors.
  • Legal requirements: We may disclose your data where required by law, regulation or court order.

We do not sell your personal data to third parties. We do not share your data with advertisers.

6. International data transfers

Some of our service providers process data outside the UK and EEA (including in the United States). Where this occurs, we ensure appropriate safeguards are in place, such as Standard Contractual Clauses approved by the UK Information Commissioner's Office or the European Commission, or reliance on an adequacy decision.

7. Data retention

We retain your personal data for as long as your account is active on the platform. If you delete your account, we will delete your personal data within 30 days, except where we are required to retain certain records for legal, regulatory or legitimate business purposes (such as records of placements made).

When we delete your data, we keep your email address, any ID from our recruitment CRM and a fingerprint of your CV on a suppression list, so that our imports and other services cannot recreate your record. We keep this list for as long as we import candidate data. It does not stop you signing up again yourself, and you can ask us to remove your entry at privacy@deciml.io.

We record the date of your last sign-in so long-inactive accounts can be identified. Where an account has seen no sign-in for 24 months, we may contact you to ask whether you wish to keep it; if we receive no response within 30 days, we may delete the account and its associated data.

Candidate records imported from our recruitment CRM that have never been used on this platform (no applications, shortlists or recruiter activity) are automatically deleted 24 months after import.

Job application records may be retained for up to 3 years after the application date for the purpose of responding to any disputes or claims related to the recruitment process.

If a company signs itself up and we decide not to approve it, we delete its account 90 days after that decision; if we have not approved it within 90 days of signing up, we delete it then. That includes the company profile, its draft roles, the accounts of the people who joined only that company, and any candidates its team added by hand. We email the company's admins 14 days before, and keep a record that the deletion happened, without anyone's name or email address.

8. Your rights

Under UK GDPR and EU GDPR (where applicable), you have the following rights:

  • Right of access: Request a copy of the personal data we hold about you.
  • Right to rectification: Request correction of inaccurate or incomplete data.
  • Right to erasure: Request deletion of your personal data (subject to legal retention requirements).
  • Right to restrict processing: Request that we limit how we use your data in certain circumstances.
  • Right to data portability: Receive your data in a structured, machine-readable format.
  • Right to object: Object to processing based on legitimate interests, including profiling for matching purposes.
  • Right to withdraw consent: Where processing is based on consent (e.g. marketing communications), you may withdraw consent at any time via your profile notification settings.
  • Right not to be subject to automated decision-making: Request human review of automated matching decisions (see Section 4).

To exercise any of these rights, contact us at privacy@deciml.io. We will respond within one month. You also have the right to lodge a complaint with the UK Information Commissioner's Office (ico.org.uk) or, if you are in the EU, your local supervisory authority.

9. Data security

We implement appropriate technical and organisational measures to protect your personal data, including:

  • Passwords are stored only in hashed form; we never store or see your password.
  • All data in transit is encrypted via HTTPS/TLS
  • Database access is restricted to authorised services only
  • API endpoints are protected by authentication and rate limiting
  • Security headers (CSP, HSTS, X-Frame-Options) are enforced on all pages
  • Access to candidate data by our recruitment team is limited to those who need it for their role

10. Children

Our platform is not intended for individuals under the age of 18. We do not knowingly collect personal data from children. If you believe a child has provided us with personal data, please contact us and we will delete it promptly.

11. Changes to this policy

We may update this Privacy Policy from time to time. Where changes are material, we will notify you by email or via a notice on the platform. The date at the top of this policy indicates when it was last updated.

For any privacy-related questions, please contact us at privacy@deciml.io.