§ 00 AI, organisations and people
I help organisations put AI to work — safely and measurably.
I am a trainer and technology executive: President of NETZ INFORMATIQUE in France, and Board Chair and Technical Manager of Reflektif, an AI-supported career guidance and competency assessment platform. I work in Turkish, French and English.
Titles are written exactly as they appear in public records.
Ecosystem map · 9 projects · 3 fields
9
live projects
5
content languages
370+
essays and articles
3
working languages · TR FR EN
§ 01 Work
Nine projects from the workshop.
Each project answers one question: how will a team use AI at work tomorrow morning — with which limits, and measured how? All of them run in the browser and are published in five languages.
A
Learn — Training programmes
The curricula I use in in-house training, complete with modules, timings and an implementation plan.
kyze.mikail.net
Corporate AI Programme
A tool-neutral, end-to-end programme for organisations: a seven-stage transformation model, a prompt library and a 90-day implementation plan. Pick one task, test it under control, measure the result.
18 modules · 47 hours · 76 templates
googleworkspace.mikail.net
Google Workspace Field Guide
End-user training from Gmail to Gemini: step-by-step instructions, hands-on exercises, a pre-test and trainer notes in one place.
15 modules · 37 hours · 7 learning paths
yzhatti.mikail.net
YZ İş Hattı — AI work scenarios
100 realistic scenarios for desk-based professionals: how the task is done today, how AI changes it step by step, and a sample prompt to try with your own data.
100 scenarios · 10 chapters · 204 sources
B
Measure & try — Assessment and awareness
Tools that make a person’s and a team’s relationship with AI visible.
kalibre.mikail.net
Calibre
A gated proficiency assessment for AI professions: choose your role, pass the level gates, and see your strengths, gaps and next step.
43 professions · 8 families · 1,290 questions
pusula.mikail.net
Pusula AI — your AI compass
A personal guide to understanding your AI habits and setting your own limits. It does not rank you; it suggests small steps for the next 7 and 30 days. No account needed to start.
8 areas · 24 statements · pilot
ses.mikail.net
Ses — just talk
A voice-only AI conversation: turn on the microphone and talk. It runs on Gemini Live; this version stores no audio, transcripts or personal data.
5 spoken languages · no recording
C
Watch — Radars and observation
Regular publications that track the field — always with sources.
yz.mikail.net
YZ Radar — daily AI top 10
A daily top-10 ranking of leading AI models and a radar of what is new — a “morning edition” refreshed every day at 08:00 (Istanbul time).
Daily · top 10 · 9 AI families
airadar.mikail.net
AI·Radar — Social Technology Radar
A research archive documenting social harms in which AI played a role — by incident, severity, institutional response and source — with a live radar and a Türkiye section.
Issue 01 · 74 records · 104 sources
skillatlas.mikail.net
Skill Atlas
Weekly skill rankings for AI agents: tools verified against independent sources, ready-made stacks, comparisons and repository scorecards. A new issue every Monday.
Weekly · 80 ranked tools · 115 repos
For the wider portfolio and services, see lekesiz.org
§ 02 Approach
An AI pilot starts with four questions.
Choosing the tool comes last. First, these four questions need a written answer; ninety days later, you measure again against the same criterion.
01
The task
Which single task are we improving? How many minutes does it take today, and who does it?
02
The data boundary
Which data may enter the model and which never will? Where do personal and confidential data stop?
03
The final sign-off
Who checks the output and who signs the decision? Where does a human stay in the loop?
04
The success measure
Which number has to change for the pilot to succeed? No pilot starts without a baseline.
Technology should amplify human potential. Public claims should remain useful, precise and independently verifiable.
Evidence policy
§ 03 About
Who is Mikail Lekesiz?
Titles are written exactly as they appear in public records.
Mikail Lekesiz is a trainer and technology executive who helps organisations put AI to work safely and measurably.
He is President of NETZ INFORMATIQUE, a company founded in 2016 and registered in France as a Qualiopi-certified training provider. As Board Chair and Technical Manager of Reflektif Bilişim A.Ş., he develops Reflektif, a platform that interprets the interests, values and competencies of students and adults with AI support but under human oversight.
In his work he insists on defining a pilot’s task, data boundary, final approver and success measure first. On mikail.net he writes about AI and people, hiring and organisational change.
Registered roles
Qualiopi
At organisation level, NETZ INFORMATIQUE: Actions de formation and Bilans de compétences · 2025–2028
France Num
NETZ INFORMATIQUE is listed as an Activateur France Num (a listing, not a label or certification)
Languages
Türkçe · Français · English
Verifiable public profile (TR) Press and speaker kit LinkedIn GitHub lekesiz.org
§ 04 Writing
Notes on AI, people and decisions.
Alongside pieces on enterprise AI and hiring, the 360° Decision and Fıtrat series ask what technology does to people. Most essays are written in Turkish.
-
Küresel Ölçekte Yapay Zeka Uzmanının Yetkinlik Profili (2026)
1 Kasım 2026 Bugün dünya ölçeğinde yapay zeka uzmanı olarak görev yapabilmek için beş katmanda donanım gerekiyor. Birinci katman temel bilgidir: matematik…
-
İnsan kaynaklarında yapay zekâ neyi hızlandırıyor, neyi kendiliğinden değiştirmiyor?
Kazanç neden sürece geçmiyor, karar neden insanda kalmalı ve endüstri mühendisleri bu tartışmada nerede duruyor Bu yazı, İTÜ Endüstri Mühendisliği Bölümü’nün Sektörden…
-
Haddini ve hududunu bilen insan
Haddini ve hududunu bilen insan Hangi sınır insanı olgunlaştırır, hangi sınır onu küçültür? Araştırma ve metin: Manus AITarih: 19 Ekim 2026Çerçeve: 360…
§ 05 Talks and press
On stage and in the classroom.
3 November 2026
ITU Industrial Engineering · “Voices from the Sector” — “Human and Intelligence” session
With Halil Kılıç. My question: AI saves time in HR — so why doesn’t the gain reach the process? If a report is written in half an hour but the approval chain still takes three days, the bottleneck hasn’t moved.
PRESS · KIT
Press and speaker kit
Approved titles, short and long biographies, portrait and usage notes. Please use these texts in event announcements.
§ 06 FAQ
Short answers.
What do you train people in?
Enterprise AI literacy and adoption, AI-assisted work with Google Workspace, secure software and cyber-awareness, and AI in hiring and competency assessment. Each programme is adapted to the organisation’s own workflow.
In which languages do you teach?
Turkish, French and English. The interfaces of the projects on this site are also available in German and Spanish.
How does an enterprise AI pilot start?
Pick one task and measure how it is done today. Write down the task, the data boundary, the final approver and the success measure. Run the pilot with a controlled group and measure again after ninety days against the same criterion.
Whom should I contact for training in France?
Training and IT services in France run through NETZ INFORMATIQUE: info@netzinformatique.fr. NETZ’s Qualiopi certificate covers “Actions de formation” and “Bilans de compétences”. For career guidance, competency assessment and institutional partnerships, contact Reflektif: direction@reflektif.net.
§ 07 Contact
Planning an AI pilot in your organisation?
For training, pilot design or speaking invitations, LinkedIn is the fastest route. For organisational requests, please use the relevant company address.
info@netzinformatique.fr · direction@reflektif.net
Privacy: no personal phone number or home address is published on this site.