What we build, and what it costs.
We build software, AI, automation, and infrastructure — productized where it can be, scoped where it should be. Every service below lists what’s included, what isn’t, and where it can go further.
Prices are starting ranges — every project is scoped and quoted in conversation. Third-party costs (LLM tokens, hosting, domains) are billed to you directly, never bundled.
Senior ownership, start to finish
A single technical lead owns delivery and pulls in specialists as needed — you're never handed off to a junior.
Fixed ranges, and you own the code
Quoted as a fixed starting range, not an open hourly meter — and source code plus handover docs come with every build.
AI-native, not retrofitted
We build and run our own AI tooling in production, so AI and automation work is core to how we operate — not a bolt-on.
Honest about what isn't ours
Third-party costs — tokens, hosting, domains — are billed to you directly, never marked up or hidden inside our quote.
Build
Custom software, applications, and web products — engineered to scope and shipped.
Software development
Bespoke business softwareCustom internal tools and business systems built around your actual workflow — not an off-the-shelf compromise.
- +Discovery & scope definition
- +Design, build, and testing
- +Deployment to your environment
- +Source code + handover documentation
- −Ongoing feature development after handover
- −Content, copywriting, or data entry
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Maintenance & support retainer (from €190/mo)
- →Additional modules or integrations
- →Team training & onboarding
App development
iOS & AndroidCross-platform mobile apps (React Native) — from a focused MVP to a full production app with backend.
- +UX flows & app build
- +Both iOS & Android (Standard+)
- +Store submission support
- +Source code + handover
- −App Store / Play Store developer account fees
- −Ongoing feature work after launch
- −Marketing assets & ASO
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Maintenance & OS-update retainer
- →Backend / API as an add-on
- →Analytics & crash-reporting setup
Web-app development
SaaS-style web applicationsValidated MVPs through to production SaaS — auth, database, dashboards, billing.
- +Auth, database, core features
- +Responsive UI build
- +Deployment & CI setup
- +Source code + handover
- −Unlimited revisions (scope is fixed per tier)
- −Native mobile apps
- −Content & copywriting
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Managed hosting & maintenance retainer
- →Payment / Stripe integration
- →Admin dashboard or reporting add-on
AI
Add AI to what you already run, or build an AI-native product from scratch.
AI integration
LLM features in your existing productindicativeAdd AI to an app you already have — chat, summarization, semantic search, RAG over your data.
- +Integration into your stack
- +Prompt design & evaluation
- +Testing & guardrails
- +Handover & docs
- −Your ongoing LLM API / token costs
- −Model fine-tuning (unless scoped)
- −Data labeling / dataset creation
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Ongoing prompt / model tuning retainer
- →Eval harness & monitoring
- →Additional AI features
AI apps
Standalone AI productsindicativeAI-powered applications built from the ground up — where the model is the product, not a bolt-on.
- +Product & AI architecture
- +Full application build
- +Evaluation & safety guardrails
- +Deployment + handover
- −Ongoing inference / API costs
- −Proprietary training data
- −Ongoing feature work after launch
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Managed operation & tuning retainer
- →Scaling & cost-optimization pass
- →Additional model / provider support
Governed AI-agent platforms
Agents that act — safelyindicativeA control plane for AI that takes real-world actions: agents propose, policy decides, a human approves, and a single gateway executes — with every action provable after the fact.
- +Policy-gated execution gateway
- +Human-in-the-loop approval flow
- +Tamper-evident audit trail
- +Deployment + handover
- −LLM inference / token costs
- −Third-party tool subscriptions
- −Ongoing policy tuning (unless retained)
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Managed agent-ops retainer
- →Additional connectors / integrations
- →Compliance & evidence-export tooling
Agents & Automation
Fast, productized automation — bots, AI agents, and self-hosted agent stacks.
Custom Python bots
Telegram · Discord · automation · scrapingSingle-purpose bots that do one job well — notifications, automation, data collection, integrations.
- +Bot build & deployment
- +Basic configuration & docs
- +One platform (Simple tier)
- −Third-party API / platform fees
- −Content moderation policy
- −Ongoing changes after delivery
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Hosting & uptime retainer (from €100/mo)
- →Extra platforms (WhatsApp, Slack…)
- →Admin dashboard
AI agents
LLM agent workflows & automationindicativeAgents that take action — multi-step workflows with tools, memory, and integrations, not just a chatbot.
- +Agent design & build
- +Tool / API integrations
- +Guardrails & testing
- +Deployment + docs
- −LLM inference / token costs
- −Third-party tool subscriptions
- −Ongoing autonomy tuning (unless retained)
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Managed agent ops retainer
- →Additional tools / integrations
- →Monitoring & eval dashboard
Hermes setup
Self-hosted open-source agent (NousResearch Hermes)Deploy and configure Hermes — the open-source, self-hosted AI agent — on your own infrastructure, connected to your model of choice.
- +Install & configuration on your infra
- +Model connection (Claude / GPT / open models)
- +Messaging-platform setup
- +Handover & runbook
- −Your server / GPU / hosting costs
- −LLM inference / API costs
- −Custom skills (Production tier and up)
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Managed ops & tuning retainer (from €500/mo)
- →Custom skill development
- →Additional platform integrations
Infrastructure
Owned, hardened server setup — from a quick deploy to a full CI/CD pipeline.
VPS & server setup
Provisioning · hardening · deploymentGet your app onto a server that's properly configured, secured, and deployable — instead of a fragile manual setup.
- +Server provisioning & deploy
- +Security hardening (Hardened+)
- +SSL, backups, basic monitoring (Hardened+)
- +Setup documentation
- −The server / hosting bill (your provider)
- −Ongoing server management (unless retained)
- −Application code changes
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Managed server & monitoring retainer
- →CI/CD pipeline as an add-on
- →Incident response / on-call
Ongoing & managed
Monthly retainers that keep what we built — or what you already run — healthy and improving.
Maintenance & support
Monthly retainerKeep your app, site, or system healthy — updates, fixes, small changes, and someone to call when something breaks.
- +Dependency & security updates
- +Bug fixes & small changes
- +Response within agreed SLA
- +Monthly status report
- −Large new features (scoped separately)
- −24/7 on-call (unless agreed)
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Faster SLA / priority tier
- →Dedicated monthly build hours
- →On-call / incident response
Managed hosting & DevOps
Monthly retainerWe run your infrastructure so you don't have to — monitoring, backups, deploys, and uptime on owned or cloud servers.
- +Uptime monitoring & alerting
- +Backups & restore testing
- +Deploys & patching
- +Monthly infra report
- −The server / cloud bill (your provider)
- −Application feature work
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →24/7 incident response
- →Cost-optimization review
- →Disaster-recovery setup
AI & agent operations
Monthly retainerindicativeKeep deployed AI agents, bots, and Hermes instances running well — monitoring, tuning, and adapting as models and needs change.
- +Agent / bot health monitoring
- +Prompt & model tuning
- +Adapting to model/provider changes
- +Monthly performance report
- −LLM inference / token costs
- −Building brand-new agents (scoped separately)
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →New skills / capabilities
- →Eval & quality dashboard
- →Additional agents under management
Audits & quick wins
Fast, fixed-price engagements — low risk, high signal. A good first step before a larger build.
Technical & AI audit
Fixed-price · ~1 weekindicativeA focused review of your codebase, infrastructure, or AI setup — what's solid, what's risky, and what to do next, in a clear written report.
- +Structured review of the target
- +Prioritized findings & risks
- +Concrete recommendations
- +Written report + walkthrough call
- −Implementing the fixes (separate engagement)
- −Ongoing monitoring
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Implementation of the findings
- →Follow-up advisory retainer
Automations & integrations
Connect your tools · custom workflowsindicativeStop doing it by hand — connect your tools and automate the repetitive work, with n8n/Make where it fits or custom code where it doesn't.
- +Workflow design & build
- +Tool / API connections
- +Testing & handover
- −Third-party tool subscriptions
- −Ongoing changes (unless retained)
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Maintenance & monitoring retainer
- →Additional workflows
- →Custom dashboard
High-assurance systems
When correctness, safety, and an auditable trail are non-negotiable — systems where the rules are enforced in code, not policy.
High-assurance & audit-grade systems
Safety-critical, compliance-readyindicativeEvent-sourced systems with cryptographically tamper-evident audit, human-in-the-loop approval, and deterministic replay — built for work where you have to prove what happened, not just claim it.
- +Tamper-evident, hash-chained audit
- +Human-in-the-loop approval & fail-closed design
- +Test-enforced safety invariants in CI
- +Deployment + handover
- −Formal certification / external audit fees
- −Ongoing compliance operations (unless retained)
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Compliance & evidence-export tooling
- →Ongoing assurance / monitoring retainer
- →Independent verification harness
Custom equipment & hardware
Hardware-integrated systems — the software, the sensors, and the custom equipment that runs them. Build-only; scoped per project.
Autonomous drones, radar & custom equipment
Hardware + software · scoped on requestindicativeWe build the software and hardware behind autonomous observation systems — flight control, sensor and radar fusion, and the custom equipment that runs them. Every consequential action stays human-authorized and fail-closed. Bespoke; scoped per project.
- +Autonomous flight & sensor/radar fusion software
- +Swappable simulator/hardware abstraction layer
- +Server-side safety governor (geofence, altitude, power, link-loss)
- +Source code + handover
- −Airframes, sensors & physical hardware (sourced to spec)
- −Flight operation, piloting, or any operational service
- −Regulatory authorization & certification
- −Third-party costs — LLM API tokens, model inference, hosting/servers, domains — are billed to you directly, not bundled.
- →Custom equipment design & integration
- →Ongoing software maintenance retainer
- →Additional sensor / platform support
Also available — scoped on request
Beyond the services above, I take on broader technical builds with the right specialists — priced per project, not from a list.
- +E-commerce & Shopify
- +Browser extensions & plugins
- +Data pipelines & scraping at scale
- +Platform & data migrations
- +API development & integration
- +Rapid MVP / prototype sprints
How an engagement works
No long sales cycle. You describe the problem; we scope it, agree a fixed range, build it, and hand it over — code and all.
- 01
Tell us the problem
Send the problem you're solving, not a finished spec. A short call or message is enough to start.
- 02
Scope & fixed range
We define what's in and out, then quote a fixed starting range — so there's no open-ended hourly meter.
- 03
Build
We design, build, and test against the agreed scope, with progress you can see — not a black box.
- 04
Handover
You get the working software, the source code, and documentation. Optional maintenance retainer if you want us to keep it healthy.
Questions, answered straight.
The things people ask before we start. If yours isn’t here, just ask.
Fixed starting ranges, not an open hourly meter. Every project is scoped in conversation and quoted as a range up front, so you know roughly where it lands before we begin. The prices on this page are those starting points.
Payment is split across agreed milestones — so you pay as verified progress lands, not everything up front. We confirm the schedule together when we scope the project.
It depends on scope, and we confirm the timeline when we quote. As a rough guide: small bots and automations land in days; an MVP is usually around 2–4 weeks; larger builds run 1–3 months or more.
You do. Source code plus handover documentation is included in every build — you're never locked into us to run or change what we made.
Third-party costs — LLM API tokens, model inference, hosting and servers, domains, and paid third-party services — are billed to you directly, never bundled into our quote. Each service above also lists exactly what is and isn't included for that work.
Noa'Lia operates as a lead-and-specialist team: a single technical lead and contractor owns delivery, bringing in the right specialists for the project. You get senior ownership end-to-end, not a junior handed your account.
Each tier is a fixed scope, so changes beyond it are agreed and quoted before we do them — no surprise additions. We'd rather re-scope honestly than quietly let the bill drift.
Revisions to meet the agreed scope are included — we fix anything that doesn't match what we signed off on. After delivery, bugs in what we built are fixed free for 30 days. New scope beyond the original is quoted separately.
Optionally, yes. Maintenance and support retainers (from €190/mo) cover updates, fixes, and small changes; managed hosting and AI/agent operations retainers are available where they fit. Handover means you can also take it fully in-house.
Yes. We're based in Ljubljana, SI · EU, and work remotely. The pricing here is Slovenia-competitive; the work isn't limited to local clients.
Tell us what you need.
Send us the problem, not a spec. We’ll scope it, quote a fixed range, and tell you honestly what’s in and out.