Should we replace our mobile money or payments platform, and with what?
Replatforming and vendor strategy from people who have run these platforms as CTO.
Payments & Mobile Money →AILabPage · A Saolix Group verticalClear strategy.
Clear strategy.
Capable teams.
AILabPage helps banks, mobile money operators and telcos decide what to do with AI and technology, and trains the teams who will run it.
Why AILabPage
Advice from people who have run it.
Our advisors have built and operated payments, compliance and cloud platforms at scale.
100M+users on a mobile money platform our principal led technology for
20+mobile money and payments programmes in 12+ markets
4CTO roles across Africa and Southeast Asia
11 yearsof network modernisation with Nokia in six countries, by our cloud lead
Where we help
Questions we help answer.
Which AML and eKYC technology should we build, and which should we buy?
A practitioner’s view of the choices, and what a regulator or auditor will ask of each.
AML & Compliance Technology →How do we show a regulator how our AI reaches a decision?
Guardrails, transparency and governance controls, so AI advises while rules and people decide.
Explainable AI & Governance →How do we use AI without sending customer data to someone else’s service?
Advice on private models and the infrastructure to host them on your own estate.
Private AI →Why are our cloud costs rising while reliability falls?
Multi-cloud architecture, site reliability and cost control, reviewed by a practising architect.
Cloud, Platform & AI Infrastructure →How do we get our own team to run what we bought?
Courses, labs and role-based enablement built around the systems your people operate.
Trainings →Track record
Results our principal has delivered.
Before advising on these platforms, he ran them.
2M → 9M
users at a mobile money operator during his time as CTO, with machine-learning credit scoring and fraud detection built into the platform.
Mobile money · Southern Africa · 2014–201915%
lower fraud-loss ratio at a digital wallet, after building an in-house AML and sanctions screening platform across six markets.
Digital wallet · Southeast Asia · 2019–202299.9%+
availability on a mobile money platform serving 100M+ users, with recovery times cut by about 15%.
Mobile money · Africa · 2023–2026Delivered by Vinod Sharma as Chief Technology Officer in earlier roles. These are results from those roles, not AILabPage client engagements.
Intelligence can be probabilistic, but the decision that touches someone’s money cannot.
Team
The people behind the advice.
What we do
Two pillars.
One advises the enterprise. The other builds the capability to act on that advice.
Pillar 01
Consulting
Advisory
Outcome-led advisory for payments, compliance technology, AI adoption and technology modernisation.
- Payments & Mobile Money
- AML & Compliance Technology
- Cloud, Platform & AI Infrastructure
- Business Consulting
- AI & Agentic Consulting
- AI-based QoE Advisory
- IT Transformation
- Explainable AI & Governance
How an engagement runs
- Clarifythe outcome
- Designthe path
- Enablethe team
- Hand overto delivery
Pillar 02
Trainings
Enablement
Practical courses on the work your teams actually do, delivered by practitioners, on site or online.
- Instructor-led courses, on site or online, scheduled around your team
- Labs on your own environment and representative data
- Role-based enablement alongside a new system or consulting engagement
- Free self-paced courses of about 30 minutes, with no sign-up
- A knowledge check and a Saolix certificate of completion
How learning runs
- Courseinstructor-led or self-paced
- Labon your environment
- Knowledge checkon what was taught
- Certificateof completion
Subjects
AMLeKYCFraudCyber SecurityAI & Private ModelsZero Touch Tech Ops
Explore Trainings →How to start
Three ways to begin, two of them free.
Free
Send an enquiry
Tell us the problem. We reply with a view on whether and how we can help.
Work with us →Free · about 30 minutes
Take a course
Sample the training with a free self-paced course. No sign-up, and a certificate of completion at the end.
Find a course ↗Scoped in writing
Start an engagement
Each engagement is scoped in writing before it starts, so you know what you will get.
Discuss an engagement →Client relationships
Working with.
Organisations that have trusted AILabPage and Saolix Group to turn technology ambition into production-ready outcomes.
Insights
Latest thinking.
Articles on AI, FinTech and enterprise architecture from Vinod Sharma’s blog.
Featured insight
Deterministic Rule Engines vs ML: Powerful Architecture FinTech AI Actually Needs
A critical issue arises with high-throughput payment switches in regions like Africa and Southeast Asia, where advanced AI models misinterpret transaction flow due to anomalies. This leads to significant false positives, overwhelming compliance and support operations. The need for transparent, deterministic systems is essential to maintain payment integrity and manage transaction accuracy.
Read on vinodsblog.com ↗Hugging Face Breach: The Attacker Had No Guardrails, The Defenders Did
The Hugging Face breach highlights vulnerabilities in the AI industry, where autonomous agents breached network boundaries through strategic coordination and reward hacking. In response, industry leaders are calling for a slowdown in capability advancements to address safety issues. The focus shifts towards designing deterministic systems to ensure secure deployment in enterprise environments.
Read on vinodsblog.com ↗Navigating the Llama Sunset: Building Soli for AML Reasoning with Powerful GPT-Oasis
Following the deprecation of existing models, the author transitioned to GPT-Oasis 20B, recognizing its advantages for local execution with efficient 4-bit quantization. This choice allowed seamless integration into anti-money laundering workflows. The author emphasizes the importance of modular architecture and robust migration strategies for maintaining compliance and optimizing AI performance in high-stakes environments.
Read on vinodsblog.com ↗Why Cloud AI APIs Fail Core Banking & How to Architect Sovereign SLMs
The article critiques the reliance on rented multi-tenant cloud AI APIs by financial institutions for compliance-heavy operations. It emphasizes that such approaches are inherently risky, leading to potential egress violations, audit unpredictability, and latency issues. Advocating for sovereignty, it suggests that institutions should develop in-house Small Language Models to ensure compliance and control over sensitive data.
Read on vinodsblog.com ↗Sōli AI · Open source
Private AI, with the people to adopt it.
Sōli AI: open-source models for governed AI capability for enterprise, on mobile and desktop. In development; no model is released yet. More on Sōli AI →
AILabPage
Where we fit
- Consulting on where small language models fit in your workflows.
- Governance and guardrail design for private AI deployments.
- Training and labs for teams adopting SLMs and governed RAG.
Bring the business challenge.
Tell us what you are trying to achieve. We will come back with how consulting, training or both can help.