Est. 2020 Data annotation & AI services

Training data you can trust.

Ba-anote, a subsidiary of Ballspedia Technologies, is the data annotation and AI services arm of the group. We build and manage trained annotation teams that label data, evaluate AI model outputs, and help search engines match user intent with results.

2013Ballspedia group founded
2020Ba-anote launched
NDAbound teams, by default

Ada wants to returnINTENT her order because it arrived a size too smallISSUE. She's asking for a refund or exchangeACTION within 14 daysTIMEFRAME.

01 Who we are

Applied AI services, built and managed from Nigeria.

Ba-anote launched in 2020 as Ballspedia Technologies' move into applied AI services. What began as a small annotation unit has grown into a managed workforce operation supporting AI labs, search platforms, and technology companies with high-quality human-labelled data.

We operate confidentially. Teams work under strict NDAs and client-specific quality protocols, which means the work is often invisible — and that's the point. Our value is accuracy, throughput, and discretion.

Built and managed from Nigeria, Ba-anote taps a large, underused pool of educated talent and gives it structured, supervised work in the AI economy.

Client workspace — access restricted
NDA signed QA sampling active
02 Services

Five ways we support your models and your search.

Data annotation & labelling

Text, image, audio, and video: classification, entity tagging, bounding boxes, transcription, sentiment labelling, delivered to client specification.

AI model training & evaluation

Trained teams that produce, rate, and rank model responses, run side-by-side comparisons, write reference answers, and flag safety or quality issues.

Search relevance

Assessing whether search results answer what the user actually meant, closing the gap between query and intent.

Managed annotation teams

We recruit, train, supervise, and quality-check annotators on your behalf, against agreed throughput targets and reporting.

Quality assurance & audit

Second-pass review of existing datasets, inter-annotator agreement scoring, and guideline refinement.

03 How we work

A five-stage process, from scope to delivery.

1
Scope

We review your guidelines, data type, and volume, and agree quality benchmarks.

2
Pilot

A small batch establishes accuracy baselines and surfaces guideline gaps.

3
Team build

We recruit and train annotators matched to the task and language requirements.

4
Production

Work runs at agreed throughput with continuous QA sampling.

Delivery & review

Data is delivered in your format, with quality reporting and a feedback loop.

04 Why Ba-anote

What working with us actually looks like.

Trained, not crowdsourced Annotators are onboarded, tested, and supervised — not anonymous gig workers.
Confidentiality by default NDA-bound teams and controlled data handling.
Contract & compliance infrastructure Formal employment letters, contract management, and a live worker portal — unusual in this sector.
Scalable capacity Teams expand for volume spikes without a drop in quality.
Backed by an established group Ba-anote sits within Ballspedia Technologies, part of an ecosystem operating since 2013.
05 Industries

Where our teams already work.

AI research labs Search engines E-commerce Fintech Healthcare Media & content platforms Academic research
06 Careers

Apply to become a data annotator.

We are always building capacity. If you have strong written English, attention to detail, and the patience to work to a guideline, we want to hear from you.

  • Applications reviewed on a rolling basis
  • Successful candidates invited to a short assessment
  • We reply within 14 working days if there's a match

All fields are required unless marked optional.

Enter your name exactly as it appears on your ID.

We will only contact you through this address. Check it carefully.

Include your country code, e.g. +234.

0 / 300

Thank you. Your application has been received.

If your profile matches an open project we will contact you within 14 working days. If you don't hear back within that window, you may reapply after three months.