Thibault Ketterer

- Audited client AI stacks and recommended cost-efficient architectures - Fine-tuned LLMs (Llama 3.2, Mistral) on domain-specific data, reducing hallucination rates by 40% via QLoRA + quantization techniques with Unsloth.

Yogosha is a pioneering French startup making its mark in the security sector. As a key member of their team, I am leading the design and implementation of their Vulnerability Operation Center (VOC) as both a service and on-premise solution. My mission is to revolutionize the existing infrastructure by transforming it into containerized applications hosted on Kubernetes, starting from scratch. I focus heavily on security while introducing numerous new features. My expertise spans: - Kubernetes - Helm - Traefik - K3s - gitalb CI/CD - Terraform - Advanced Encryption Techniques - Private Docker Repositories - Ensuring Reproducibility and Low Maintenance with Self-healing Services - Python Scripting for Automation and Efficiency I successfully re-engineered the deployment process of the main application, reducing setup time from six months to mere seconds. Following this, I integrated all existing services into a Kubernetes cluster, streamlining operations and enhancing scalability and monitoring / logging.

I am passionate about blockchain technology and have been an active investor since 2018. My journey in the blockchain space includes: - Extensive Research: I have read over 300 whitepapers to stay informed about emerging trends and technologies in the blockchain ecosystem. - Project Testing: Regularly testing new projects and using DeFi platforms to understand and evaluate their functionalities. - Token Creation: Developed and deployed tokens for testing purposes to explore different blockchain functionalities. - Smart Contract Development: Created and tested various smart contracts to ensure their security and efficiency. My expertise includes: - Cryptocurrency Mining: Proficient in mining digital currencies, including Ethereum, Monero, and Bitcoin. - Blockchain Testnet Specialist: Actively participating in and supporting various blockchain testnets such as Aptos, Namada, Celestia, Entangle, Dymension, Avail, Taiko, and Quilibrium. I assist in the development and testing of new blockchains, identify bugs, contribute code improvements, and report security vulnerabilities. I am deeply committed to advancing blockchain technology and continuously exploring new opportunities within the space.

Consulting Expertise: I offer consulting services in the following areas: - Web Tracking and Analytics: Implementing and optimizing tracking systems for insightful data analysis. - High Traffic Websites: Specializing in caching and bench-marking to ensure optimal performance and scalability. - Backup Strategy: Designing and implementing robust backup solutions for data security and disaster recovery. - Technical Architecture: Crafting efficient and salable technical architectures tailored to business needs. Clustering, High Availability (HA), and Load Balancing: Ensuring system reliability and performance through advanced clustering and load balancing techniques. - Cloud Hosting: Expertise in AWS, Google Cloud, and Terraform for seamless cloud integration and management. - Automation: Utilizing Ansible and SaltStack for efficient and error-free automation of infrastructure and processes. - VPN Implementation: Securing network communications with robust VPN solutions. Debugging and Problem Solving: Diagnosing and resolving complex technical issues swiftly and effectively. My consulting services are designed to provide comprehensive solutions, ensuring your systems are efficient, secure, and scalable.

Data Infrastructure Team at Bedrock Bedrock is building streaming platforms (like netflix) the most know customers are currently Salto (new french SVOD platform) and 6play(M6 TV channel) I am the Data infrastructure needed to handle SVOD data (medias and user tracking) aws + EMR for data platform airflow as a scheduler, dataviz by Superset. From 1 to 10 customers in one year. We migrate from a POC to a complete EKS managed solution, scalabe and reliable, monitoring with prometheus and aws cloudwatch, some PRs to superset. We manage to transform and orchestrate the infrastructure to scale more and multiply it for every customer, leveraging a lot of terraform. Passing from a 2 hour deployment to 2 seconds (in k8s) Tech used: airflow, python, superset, aws, emr, terraform, docker, kubernetes, jenkins with a drop if agiles processes, prometheus, grafana, eks

Part of the Core Data Team at M6web We are designing the Data infrastructure needed handle SVOD data (video or user related) using a lot of aws EMR and airflow scheduler. end of 2019: migration to EMR ephmeral hadoop clusters by aws. I used my knowledge on airflow and hadoop to help the migration to EMR(aws branded auto-hadoop) Tech used: airflow, python, superset, aws, emr, terraform, docker, kubernetes, jenkins with a drop if agiles processes

Developed and launched the complete DevOps curriculum from scratch, covering Development, Continuous Integration (CI), Cloud, Kubernetes, Testing, and Automation. Responsible for teacher recruitment and student management. Delivered comprehensive DevOps courses.

We started at 30 web servers and there were 100 when I left (mainly apache / nginx) backed up by CDN (CDNetworks and varnish) I took great care of the production and internals servers and managed the IT department (3 people), be pro-active on IT innovations, monitoring etc.. Distributing video content, sometime live event (I remember well the placebo concert in Paris)m and media content (like Loreal assets) 100 debian /apache / mysql servers serving hundreds clients websites (mainly video streaming, and live streaming) + internals preprod ... 30 host ffmeg (video encoding farm) CISCO, VLANs, and ADSL load balancing. Technologies I used: DNS, apache2, varnish, php, mysql(master-slave), OVH, git, python, bash scripts, postfix, exim, CDN, zabbix, munin, apachebench, etckeeper, fabric. I have written many scripts to interact with APIs : OVH, googledoc, cdnetworks in order to simplify my job. Many virtual machines under kvm, xen. 2013 We entered the cloud world by the MS azure door, lots of automation with azure-cli We worked a lot with customers like Loreal and INRA to secure their Digital Asset Managment services (DAM) 2012 I designed our new load balanced web infrastructure based on VLAN, cisco ACE and some servers I have done a lot of varnish-cache tuning to optimise hit rates. discovering azure Server provisioning, monitoring, storage (backing up TB of videos), tested aws and azure cloud.

built a GUI for a distributed computing in wx-perl lots of play with proxy, squid and automation with scripts

Contractor: I've made a CMS, that was used for 4 sites in php / mysql / javascript

- Built a bootsable iso of Linux, starting an automated video streaming device tech: Xfree/xorg, touchscreen, Gtk programming, Linux From Scratch, shell scripting http://www.videospot.com/index-en.html - Sent SMS in java (spam / ad service) AT programming, mysql, php