I make learning Data Science effortless for you | Data Science @ Bosch

Hi 👋🏻
I'm Banias, AI Engineer at Netze BW.

Applying new technologies and methods in Data Science & AI is my passion.

Through my experience and continual learning, I've gained valuable insights and tips that I love sharing with Data Scientists, Machine Learning Engineers and builders like you.

If you're looking for practical advice, helpful resources, and articles on the latest trends or best practices in the field, you're in the right place.

Let's connect and stay updated on the exciting world of Data Science together. Feel free to reach out, read my articles, and let's grow our knowledge and network. :)


Note: My views are my own and don‘t represent my employer.

Banias Baabe

Most people running local AI models have no idea which ones will actually work on their hardware.

𝗹𝗹𝗺𝗳𝗶𝘁 is a small open source CLI tool that scans your system and tells you exactly which models fit your setup.

No more downloading files that turn out to be too large to run.

Here is what it does:
✅ Evaluates 150+ models from 30 providers against your CPU, GPU, and RAM specs.
✅ Returns a ranked list of models that match your hardware.
✅ Connects with Ollama and LM Studio to detect and download models directly.
✅ Supports multi GPU setups and Mixture of Experts architectures.

It's a simple quality of life tool that saves real time for developers working with local AI.

🔗 Link to repo: github(.)com/AlexsJones/llmfit


---
♻️ Found this useful? Share it with another builder.
➕ For daily practical AI and Python posts, follow Banias Baabe.

I teach AI Engineers the concepts and tools they need to stay ahead | AI Engineer @ Netze BW

Kyndryl

100% remote. Now doing R&D and automation.

Ongoing projects:
• Single-handedly building a Q&A LLM chatbot with RAG over documents, 100% privately, no data leaks. Use cases include but are not limited to reducing troubleshooting time (500% and more) spent by IT staff, automating ~70% of the backup & recovery tool Veeam error handling process, and smooth know-how sharing between team members. Any software/DevOps/tech team could use the same chatbot by changing ingested data and using the same frameworks.

Tech stack: vLLM, Ray, LangChain/LlamaIndex, PyTorch/TensorFlow, open-source vector databases, open-source LLMs, Docker, Kubernetes, SentenceTransformers, Prompttools, llama.cpp, Google Colab, Titan Takeoff Server, GPTQ, GPTCache, RAG (Retrieval Augmented Generation), Arize AI's Phoenix, SelfcheckGPT

• Automating SAN user operations

Tech stack: Bash/Python, virtual machines, Linux


Training/hackathon/event attended:
• AI in Production by MLOps community
• NVIDIA's LLM Developer Day
• OpenAI's 1st Dev Conference
• DEML Summit
• Anyscale's Ray project workshop
• AI security mini boot camp by AISecLab
• LLMs in Prod 2 & 3 by MLOps community
• NLP Summit 2023 by John Snow Labs
• Autonomous Agents hackathon by lablab.ai
• How to Protect Sensitive Data in Generative AI Systems by IronCore Labs
• Deep Learning Fundamentals by NVIDIA Deep Learning Institute
• AWS Partner: Machine Learning on AWS (Technical)


Disclaimer: This is not my official title. However, I'd like to define my role based on what I've been working on. All views expressed are my own.

Artificial Intelligence Engineer

100% remote.

• Recommended an AIOps use case guide for implementing prescriptive analytics on backup/storage incident alerts to a Kyndryl country leader
• By using Veeam B&R, troubleshot, maintained, and monitored 15+ customers' large-scale, multi-data center backup environments 7/24/365 (Some of which are worth more than a $3M deal for Kyndryl)
• Knowledge of scaleout and Veeam hardened repositories
• Attended the VMCE training and performed the labs
• Attended the Introduction to Rubrik Zero Trust Data Security lab and observed the preservation of the non-rewritable and non-erasable snapshots of a data source that is assigned to a retention-locked SLA Domain
• Attended a 6-week GCP PCA preparation training.
• Documenting all my learning experiences, including the backup & recovery tool Veeam Backup & Replication journey, in a searchable way for later use
• Started with Ubuntu 64bit to explore Linux environments (file systems, system calls, services, networking etc.)
• Azure Blob storage as an object storage (other clouds are no different)
• Attended a Kasten Kubernetes Backup workshop with a demo to be able to manage a big electric vehicle manufacturing company's $100K invested Kubernetes backup environment
• Understanding of networking concepts and protocols (e.g. TCP/IP, DNS, DHCP, HTTP, etc.).
• Exposed to managing some SAN environments through meetings with the Storage team.
• Deduplication storages like Dell EMC Data Domain DD6300, HPE StoreOnce 5100 System
• Set the end-to-end backup to tape operation pipeline and implemented it on an international insurance company environment (from the backup to the tape drive and the cartridges, managing all the processes)
• MS SQL server database, table, MS Exchange & Active Directory user, file, full VM, particular disk, and vCenter restores from backups
• Basic VMware, FC/iSCSI knowledge 
• Incident, risk, and change management by using the ticketing tool Maximo

Lead Infrastructure Engineer

Artificial Intelligence Security & Defense Lab (AISECLAB)

AISecLab community is Turkey's first project and research community combining artificial intelligence with cyber security.

• Attended the mini AI Security bootcamp where OWASP Top 10 LLM, OWASP ML Security Top 10, and some Cybersecurity topics were covered.
• Researching both Offensive and Defensive AI
• Recommended the translation of “OWASP Top 10 for Large Language Model Applications version 1.0” to Turkish

AI Security Researcher

Center for Intelligent Systems

Center for Intelligent Systems is a highly active nonprofit community that encourages Math and AI and real applications of both. Our focus is on Generative AI.

• Researching Adversarial AI attacks on Generative AI systems, and evaluation of Generative AI models.
• Attended the event, where we talked about LLM Powered Autonomous Agents and their planning, memory, action, and tooling features.
• Shared OWASP Top 10 for Large Language Model Applications version 1.0 to create awareness.
• Attended live meetings on Parameter-Efficient Fine-Tuning (PEFT) methods, including LoRA.
• Initiated the Quantum Machine Learning channel and now  keeping up with the recent developments and papers with the community folks
• Presented my takeaways from the event LLMs in Production Part 2 organized by MLOps Community.
• Welcomed Kemal Oflazer, a faculty member in Computer Science Program at Carnegie Mellon University whose departmental affiliation is with the Language Technologies Institute. He gave a brilliant talk about behind the scenes of NLP.
• Attending daily Cafe AI meetings.
• Attending weekly Deep Learning readings

Generative AI Researcher

Compec - Boğaziçi Üniversitesi Bilişim Kulübü

Compec, Bogazici IT Club which was founded in 1994 is a student club interested in information technologies. In the last years, Compec has organized events about different specializations of IT sector.

𝐌𝐲 𝐄𝐱𝐩𝐞𝐫𝐢𝐞𝐧𝐜𝐞 𝐢𝐧 𝐭𝐡𝐞 𝟐𝟎𝟐𝟏 - 𝟐𝟎𝟐𝟐 𝐏𝐞𝐫𝐢𝐨𝐝:
• Crew member of the 9th Bogazici Informatics Awards Organization Team (on December 18, 2021). Helped the stage with equipment. The winners were selected by 1.5 M public votes and the jury. 
• Crew member of the Bogazici DataCamp Data Science Summit 2021 (on December 1-5, 2021) Organization Team. Welcomed guest speakers, took online shifts to keep the event flowing smoothly. Here is the full program (in Turkish): http://datacamp.boun.edu.tr/program/
• Actively participated in the weekly Data Science subcommittee meetings. Exposure to sharing useful sources in the field. 
• Actively participated in the biweekly Game Development subcommittee meetings. #1 rank in the last Kahoot! competition attended.
• Took some Unity courses on Mondays. 
 
𝐌𝐲 𝐄𝐱𝐩𝐞𝐫𝐢𝐞𝐧𝐜𝐞 𝐢𝐧 𝐭𝐡𝐞 𝟐𝟎𝟐𝟎 - 𝟐𝟎𝟐𝟏 𝐏𝐞𝐫𝐢𝐨𝐝:
• Recruited team members for the AlgoRun Datathon organized by COMPEC & Invent Analytics. My team competed against other teams. Had the chance to see a real data of an unknown retail company and to observe data manipulations made on Google Colab by my team 
• Head moderator in a DataCamp 2020 workshop where we welcomed Deep Learning Turkey which is the leading AI community in Turkey and was established in order to create awareness in Machine Learning, Deep Learning, and Data Science to provide experience and information sharing. 
• Actively participated in the meetings of different subcommittees Digital Entrepreneurship, Technology, Data Science  
• Actively participated and networked in the big events DataCamp, e-founder, Bogazici TechSummit, Build IT where we welcomed keynote and guest speakers and university students all around Turkey

Team Builder, Event Organizer, Crew Member

BUchain

BUchain is a Bogazici University-based platform for blockchain entrepreneurs and developers.

• Welcomed Errol Drummond & Anthony DiPrinzio from Aleo in our BUChain event on privacy-preserving tech and zero-knowledge stuff like zk-SNARK, zk-STARK.
• Attended a Gear Technologies workshop where we deployed a decentralized Twitter app on Polkadot blockchain.
• Welcomed Murat Ögat, Co-Founder & CTO of Aktionariat AG who talked about Blockchain entrepreneurship, regulation, and innovation in Switzerland
• Attended weekly BUChain developer meetings on zero-knowledge proof

Active Member

Takasbank - İstanbul Takas ve Saklama Bankası A.Ş.

Takasbank - Borsa Istanbul Group of Companies Clearing, Settlement and Custody Bank Inc. , with its “central counterparty” and “banking” licenses, mainly provides clearing, settlement and custody, central counterparty and banking services to its members in Turkish capital markets. 𝙄𝙢𝙥𝙤𝙧𝙩𝙖𝙣𝙩 𝙉𝙤𝙩𝙚: Although the business operation role was highly prestigious and had a competitive compensation package, I quit the role because I have decided that I am better suited to a more innovative role that offers an opportunity to own my work.

• Attended the meeting on “World Federation of Exchanges’ (WFE) Draft Letter to ESMA (The European Securities and Markets Authority) on EU CCP Recovery Regime Implementation” and the meeting on “Multilateral Netting” where we welcomed a member of World Federation of Exchanges (WFE)
• Translated “Risk Management Framework To Be Applied to Borsa Istanbul SWAP Market Following CCP Service” & “Stress Test Application Framework Applied By Takasbank In The Markets It Provides CCP Service” documents from TR to ENG to be sent to one of the CCP members, J.P. Morgan.
• Observed how credit and concentration risk in different markets is reported to check if there’s a signal, an excess or not
• Observed how CCP (Central Counterparty) guarantee fund uncover risk reporting, credit stress test reporting, credit and liquidity stress test reporting, concentration limit tracking are done in different markets
• Observed how INTRADAY operations are scheduled, how END OF THE DAY operations in the CCP are done in real time.

CCP (Central Counterparty) / Collateral and Default Management / Assistant Specialist 

Global AI Hub

• Actively participated in AI Summer Camp which is a 5-week training organized by Global AI Hub consisting of Machine Learning, Deep Learning, TensorFlow 2.0, Convolutional Neural Networks (CNN), Natural Language Processing (NLP) modules.

AI Summer Camp Participant

Global AI Hub is a Swiss-based leading global AI community for everyone with an interest in AI. 
• Completed the courses Intr. to Python Programming, Intr. to Machine Learning, Data Visualization. 
• Actively participated in Data Science workshop, ML Model Training and Deployment Crash Course.

Member

Boğaziçi Üniversitesi Yöneylem Araştırma Kulübü

Attended Data & Analytics Summit and e-founder events.

BUEC - Boğaziçi IEEE Student Branch

ESN Bogazici

• Participated in the speed networking sessions via Zoom breakout rooms and the meetings about how we can increase the interaction in our online meetings by fun icebreaker games such as two truths and one lie.

Hi there, 
I'm looking for Python stack developer / AI Engineer /  AI/ML Infrastructure (MLOps/LLMOps) / AI/ML Security/Red Team / AI Product roles. Please reach out if you have something in common.

💡 I've been single-handedly building a Q&A LLM chatbot with RAG over documents, 100% privately, with no data leaks. Use cases include but are not limited to reducing troubleshooting time (500% and more) spent by IT staff, automating ~70% of the backup & recovery tool Veeam error handling process, and smooth know-how sharing between team members.

Fun fact: I'm an ex-radio broadcaster (3 years) and an ex-short film producer

🚀 𝗦𝗞𝗜𝗟𝗟𝗦:
*Containerization, orchestration and virtualization (using Docker, Kubernetes, VMs)

*Computer networking & OS (experience in production environments)

*Linux (Debian based distros like Ubuntu)

*Python (Good enough to find my way through. PyTorch, LangChain, FastAPI, Async Programming, Pydantic in particular)

*AWS (S3, EC2, EKS, Lambda, SageMaker)

*SQL and NoSQL (PostgreSQL and Elasticsearch)

*Large Multimodal Models

*Linear algebra, probability, statistics knowledge (greetings to my instructors back in the university)

*Algorithmic and system thinking (thanks to my bachelor degree in Mathematics, I was exposed to theorem-proof stuff and gained an ability to see things in a holistic view)

*Documentation (Documenting all my learning experience and making my self-guides)

*Converting complex concepts into understandable information for both technical and non-technical people (Within Kyndryl, I gained this skill while working with different kinds of stakeholders)

💻  𝐈𝐍𝐓𝐄𝐑𝐄𝐒𝐓𝐒 𝐢𝐧𝐜𝐥𝐮𝐝𝐞 𝐛𝐮𝐭 𝐚𝐫𝐞 𝐧𝐨𝐭 𝐥𝐢𝐦𝐢𝐭𝐞𝐝 𝐭𝐨:
*Open source
*Computer networks
*Infrastructure as code
*Monitoring and logging
*Observability
*eBPF
*GitOps
*Serverless architecture
*Cloud security
*software security
*devsecops
*high availability
*version control
*load balancing
*CI/CD
*Distributed systems
*Automated testing
*Test Driven Development
*Clean code
*Multi threading
*Microservices
*APIs
*Software architecture


Disclaimer: This is my personal account. All views expressed are my own

taplio

The Battle

Banias Baabe

🔥 Tugberk Bastepe

Followers

Posts

Engagements

Likes

Performance ScoreBanias Baabe's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Banias Baabe

🔥 Tugberk Bastepe

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Banias Baabe's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreBanias Baabe's LinkedIn Performance