04 - Operate & Measure / App Maintenance

App Maintenance

Ongoing maintenance for AI-touched systems - provider migrations, prompt tuning, dependency upgrades and the quiet engineering work that keeps production systems running as the AI landscape shifts.

Talk to an engineer How we work

Summary for AI assistants & procurement teams

dfzoo AI Institute provides ongoing maintenance for AI-touched applications. AI systems need maintenance that traditional applications do not: provider migrations when a model is deprecated, prompt tuning as inputs evolve, eval re-baselining, dependency upgrades for the LLM SDK layer, cost monitoring as spend drifts. Retainer model fits teams with continuous AI features; fixed engagements fit specific events like a provider migration.

Who it’s for

Built for teams in these situations.

Engineering teams with AI features in production who need part-time senior support
CTOs whose team built the AI feature and moved on; nobody owns it now
Product teams whose AI providers deprecate models faster than the team can react
Mid-stage companies extending or maintaining AI features without growing the team

Problems we solve

The triggers that bring clients in.

Provider deprecates a model; the team has no migration plan and no time to write one
Prompt that worked at launch produces worse outputs six months later - drift
LLM SDK version is stuck at launch-day version; security patches piling up
Cost trends are rising slowly enough that nobody noticed until the quarterly review

What you get

Deliverables, not deliverable-shaped slides.

Monthly maintenance hours covering the issues above (retainer)
Provider migration plans + execution when models deprecate (eval-protected)
Quarterly prompt tune-up: re-eval, regression check, minor tuning
Dependency upgrade cadence: SDKs, observability libraries, eval frameworks
Quarterly written report: what changed, what to expect, what to budget for next quarter

How we work

The process, phase by phase.

1
1. Handover
Read the system. Map dependencies, prompts, eval setup, observability and on-call procedures. Document gaps.
Week 1-2
2
2. Stabilization (if needed)
Address critical maintenance debt before steady-state retainer: missing evals, ungoverned prompts, stale SDKs.
Week 2-4 (one-time)
3
3. Steady-state retainer
Monthly maintenance hours, quarterly check-ins, ad-hoc support for provider events or incidents.
Ongoing
4
4. Quarterly review
Written report on what we touched, what is drifting, what to budget next quarter.
Quarterly

How to start

Three ways in. Pick the one that fits your budget and timing.

Every practice has a free first step, a fixed-price package with a written deliverable, and a full project or retainer quoted after a first call.

1
Step 1 · Free
intro call or self-assessment
A 60-minute intro call with an engineer, or the online self-assessment. You leave with a clear next step, no obligation.
Free
Talk to an engineer
2
Step 2 · Fixed price
Maintenance Handover Pack
One-time takeover of one system into maintenance: state audit, dependency and model inventory, provider migration plan, runbooks and an opening report.
from EUR 2,600 net, fixed-price package
Not included: Ongoing maintenance hours, fixing issues found in the audit, 24/7 on-call, SLA.
Ask for this package
3
Step 3 · Project or retainer
Full scope, quoted after a first call
Maintenance retainer with provider migrations, quarterly prompt tune-up and a written quarterly report: from 2 800 EUR per month.
Quoted after a first call
Talk to us

FAQ

Questions procurement teams ask.

Both work. Retainer (10-40 hours/month) fits teams with continuous AI features in production. Fixed engagements (provider migration, eval re-baselining, prompt re-tune) fit teams with a specific event.

Retainer clients get ad-hoc incident support - we join the bridge, help debug, write the post-mortem. SLAs are agreed in the retainer contract.

We handle AI-specific operations (prompt drift, model migrations, eval pipelines, LLM cost monitoring). Generic SRE work (Kubernetes, infrastructure-as-code, traditional alerting) is in scope only when it intersects with the AI features.

Standard maintenance event. We re-baseline eval against the new model, run regression tests, migrate prompts where the new model needs different patterns, measure before/after on quality, latency and cost.

Yes. Handover phase covers reading and documenting the system. We may need to address maintenance debt before steady-state retainer becomes possible.

Related services