Scale
It exists. Now it has to hold.
Reliability, cost, and the senior judgement that stops a system becoming the thing that has to be replaced.
Software that works at ten customers fails in specific, predictable ways at ten thousand. So does software nobody has reviewed before it takes real money. This lane is the work that only gets harder if it is done twice: reliability, performance, cost, and a second pair of senior eyes before the expensive decision.
You are in the right lane if: It is live, and something about it is not safe to leave alone.
- 01Something broke and your customers noticed before you did.
- 02Your site, domain, email or M-Pesa link is down right now.
- 03The bill is growing faster than the usage.
- 04You have engineers in-house but need senior outside help on one specific build.
- 05An AI feature is in production and nobody can tell whether a change made it better or worse.
- 06Payments, settlement or fraud detection that has to actually work at scale.
- 07You want architecture reviewed before a critical build, not after.
- 08You need senior engineering leadership without committing to a full-time hire.
Reliability and incident work
Finding why it failed, fixing the class of problem rather than the instance, and leaving behind the monitoring that means you hear about the next one before a customer does.
Performance and cost control
Where the time goes and where the money goes, measured rather than guessed. Usually a small number of queries, a retry loop, or an AI call nobody capped.
Review before the expensive decision
Pre-build architecture review, security and readiness checks before real customers touch something, and second opinions on stack and infrastructure choices. Hourly or short engagement.
Fintech and production AI systems
Payments rails, settlement, reconciliation and fraud detection built for regulated environments and real money. Evaluation, observability, provider routing and cost control for AI already in production.
Engineering leadership
Fractional senior engineering for teams without a head of engineering: hiring loops, technical strategy, delivery discipline.
Ship Safe
Code/security review, bug rescue, and emergency triage for AI-built apps or broken sites: find launch blockers, fix the stuck workflow, and get a clear go or no-go on shipping.
Priced after a brief conversation
- Full read of your AI-built app or repo by a senior engineer
- Security sanity check: auth, permissions, exposed secrets, API leaks, CORS, and input handling
- Debug the stuck workflow your AI coding tool cannot get stable
- Emergency triage for crashed sites, failed deployments, or dead integrations
- Launch-risk report: blockers to fix now, items to fix before growth, debt to schedule
- Small critical fixes included; larger rebuilds scoped separately
- Plain-English handover so you know exactly what changed and what to watch next
- Hourly
- From KES 5,000 or USD 50 per hour, for one clear problem that does not need a package.
- Project engagements
- Priced by scope after a brief and a scoping call. You get the number in writing before anything starts.
- Always On
- Monthly managed service. We run the boring parts so you run the business, with priority response when something breaks. Available on any lane once we have shipped something together.
You can engage me directly or contract with Noria Technologies LTD, the practice I have run since 2020.
Systems behind this lane that are running today. Every figure is explained in its writeup.
- In adoption2026
Noria Pay
One payments service every product calls, instead of each one wiring up the providers itself.
- 4
- payment providers behind one API
- 21
- codebases that stop rewiring them
Read the writeup
- Live2026
Zana
Billing, collections and reconciliation for any business that invoices the same customers every cycle.
- 5M
- connections it can bill
- 8
- kinds of business on one ledger
Read the writeup
- Live2022
SiteBooks
Construction management for contractors, and for the diaspora clients funding a build they cannot visit.
- 100+
- construction projects run on it
- KES 50M+
- processed through the platform
- 4 years
- in continuous production
Read the writeup
- 01Send a brief. If something is down right now, say so and lead with that.
- 02We scope it on a call. For an outage that call happens the same day where I can.
- 03I send a written scope and a price within 48 hours. Triage can start before the scope is signed.
- 04Delivery is typically 4 to 12 weeks, with an optional monthly retainer once we have shipped together.
If the thing does not exist yet, this lane is premature and you will pay for rigour you cannot use. If it exists and works fine but people are doing repetitive work around it, the money is better spent elsewhere.
Email: [email protected]. The person who scopes your work is the person who builds it.