« I had the opportunity to work with Florian for over a year. During that time, as SQL SRE lead, he was invaluable in helping me grow in my engagements. He was always patient, didactic and available. I highly recommend. »
On assignment · open to new engagements
Database Reliability
augmented by AI.
Florian Courtain, freelance SQL Server DBA. 15 years across mission-critical 24/7 environments: audit, AlwaysOn high availability, performance tuning. Early adopter of DBR workflows augmented by Claude API and Claude Code, in production.
Scope an engagement Download resume
15+
years on critical environments
500+
servers · 3 datacenters
4 TB
largest SQL database
9 yrs
100% remote
Three pillars
Expertise at the intersection of three worlds.
Fifteen years on mission-critical environments, plus a modern stance on AI augmentation. Not a specialist: a complete operator.
01
SQL Server & HA
High-availability architectures, deep performance tuning, archival frameworks. Databases up to several TB, 24/7 instances.
- AlwaysOn Availability Groups
- Failover Cluster Instance
- Azure SQL Managed Instance
- SSAS · SSIS · SSRS
- TempDB / query optimisation
02
Infrastructure & IaC
Industrialised IT operations with Ansible, PowerShell DSC, Git/GitLab. SRE posture, not firefighter.
- Ansible playbooks
- PowerShell DSC
- Azure Automation Account
- VMware · Linux · Windows
- F5 · Palo Alto · CheckPoint
03 · differentiator
AI-Augmented Engineering
Operational workflows structured around the Claude API and Claude Code. Live in client engagements. Measurable.
- Claude API in production automation
- Claude Code on engagements
- Advanced prompt engineering
- AI-assisted architecture design
- Repetitive-task elimination
The real differentiator AI in production,
AI in production,
not in demos.
AI augmentation is not a marketing promise on my engagements. It is a structuring operational tool, embedded in the daily run, measurable.
- 3 to 5× faster design HA architecture, T-SQL schemas, Ansible runbooks: pair-programming iterations with Claude Code.
- Argued technical decisions Execution plan comparisons, indexing choices: AI challenges options before human validation.
- Living documentation Runbooks, post-mortems, schemas: generated alongside the deliverable. Your teams inherit a doc that breathes.
- Lower total engagement cost Repetitive tasks automated: billed time concentrated on decisions and critical execution.
# Prod query at 14s P95, unstable execution plan.
# 80M rows, date + client_id predicate, plan switch seek/scan.
$ claude "analyse execution plan + propose fix"
→ Diagnosis: parameter sniffing. Skewed bound parameter,
>99% of rows for client_id=42 vs <0.1% for the
others. Plan compiled on 1st call = scan, catastrophic
for normal-volume clients.
→ Stats histogram: 0 explicit step for 80% of client_ids.
AUTO_UPDATE_STATS_ASYNC on but 20% threshold not hit
on partitioned table (5 years history).
→ Proposal: OPTION (RECOMPILE) on the sproc + filtered
stats on top-5 volume clients + UPDATE STATISTICS
WITH FULLSCAN in the weekly maintenance plan.
# [human review of diff + replica test]
$ git commit -m "fix(perf): RECOMPILE + filtered stats invoice_lookup"
$ git commit -m "docs: parameter sniffing, pattern + KB"
✓ P95: 14s → 320ms · plan variance -97%
Selected engagements
Four grounds, four scales.
A selection of the most representative projects, described in figures and verbs.
La Française Group · HA architecture
Wearing several hats (DBA, systems engineer, network engineer) on a multi-site HA architecture: ~20 SQL Server instances (databases up to 4 TB) + NoSQL MarkLogic HA cluster, across 500+ servers in 3 triangulated datacenters. Full IaC industrialisation via PowerShell DSC orchestrated by Ansible. F5 and Palo Alto integration on the network perimeter.
ProBTP · audit & performance tuning
SQL Server estate: ~20 active servers, ~30 extended, databases up to 2 TB. Full audit, critical query optimisation, industrialised maintenance plans.
Groupe Adéquat · archival framework
Stored procedures rebuilding a 1.5 TB database in pre-prod: full DDL, referential integrity, block-by-block functional history. ~15 servers + Azure SQL Managed Instance.
Amadeus · Lead Service Reliability
~60 mission-critical 24/7 SQL Server instances, databases up to 2 TB. L1 coordination, idempotent Ansible automation for cluster deployment, alert tuning, technical leadership.
Recommendations
Reviews published on Malt
Public testimonials from professionals I have collaborated with, available on my Malt profile.
« Having crossed paths with Florian on an engagement, I can only recommend his availability, his proposal-driven attitude and his technical skills. His SQL expertise combined with PowerShell (and Ansible) mastery and security best practices make him an asset for administration and infrastructure recovery. I recommend without hesitation. »
« Florian is a seasoned DBA who has already proven his skills in administration and scripting. He is also very comfortable with network administration and security. I highly recommend him. »
First call on the house
A concrete need?
30 minutes to scope the requirement. Reply within the day. No sales pitch, just the right questions. Mission-critical engagements: €1,000-1,400/day depending on scope and engagement length.
To be straight: my hands-on expertise is SQL Server. On the other SQL databases (Postgres, Oracle, MySQL, Azure SQL), I have very little operational experience but I am open to learning. In that case, the day rate adjusts down, only fair. On pure NoSQL (Mongo, Cassandra), I would rather point you to someone who specializes in it. Better for you, better for me.
Reply within the day. On Teams, an "Away" status signals an ongoing engagement (the message comes through anyway).