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

Running a 70B LLM on a single 4GB GPU is possible with 𝗔𝗶𝗿𝗟𝗟𝗠.

𝗔𝗶𝗿𝗟𝗟𝗠 optimizes inference memory usage so you can run big models without quantization, distillation, or pruning.

It runs inference layer by layer, loading only the layer that is needed for the next compute step, then freeing that memory before moving on.​

To make that practical, it first preprocesses the original Hugging Face weights and re-shards them by transformer layer (instead of giant multi‑GB shards), so each load is focused and smaller.​

This matters because disk I/O becomes the bottleneck when you stream weights.

The main thing that still sits in GPU memory is the KV cache for attention, which is much smaller than full model weights for typical context lengths.​

𝗔𝗶𝗿𝗟𝗟𝗠 also adds prefetching so it can overlap loading the next chunk of weights with the current compute, which helps claw back some speed.

And if you want more throughput, it supports an optional “model compression” mode using block-wise quantization of weights to shrink what needs to be loaded from disk, with the project claiming up to ~3x speedup and minimal accuracy loss.

🔗 Link to repo: github(.)com/lyogavin/airllm


---
♻️ 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

Stop destroying your documents with 512-token chunks.

Traditional RAG is breaking your tables, splitting your diagrams, and mixing up your model specifications.

I just read a great article from Alex Vesa about a working data modeling approach for complex technical documents.

The problem? Most RAG systems treat a 500-page service manual like a blog post. They chunk blindly. A table that spans three pages gets shredded into meaningless fragments.

Here's what actually works:

✅ Pages, not chunks
Use full pages as your retrieval unit. That's how humans read technical docs anyway.
✅ Three-page sliding windows
Process each page with its neighbors (N-1, N, N+1). Now your system knows when content continues across boundaries.
✅ Treat tables like they matter
Extract structure. Generate metadata. Flatten for search. Keep all three representations.
✅ Multi-vector indexing
Dense embeddings for semantic search. Sparse vectors for exact matches. ColBERT for precision. Combine them.
✅ Domain-specific metadata
Model applicability, component types, cross-page relationships. Not just title and date.

Link ➙ https://lnkd.in/dJ9aJKAB

Thinking about using 𝗽𝗴𝘃𝗲𝗰𝘁𝗼𝗿 for your next project?

You might want to think again.

Here's the reality of using it in production:
1️⃣ 𝗜𝗩𝗙𝗙𝗹𝗮𝘁 𝗶𝗻𝗱𝗲𝘅𝗲𝘀 𝗻𝗲𝗲𝗱 𝗰𝗼𝗻𝘀𝘁𝗮𝗻𝘁 𝗿𝗲𝗯𝘂𝗶𝗹𝗱𝘀. New vectors throw off your cluster distribution over time.
2️⃣ 𝗛𝗡𝗦𝗪 𝗶𝗻𝗱𝗲𝘅𝗲𝘀 𝗮𝗿𝗲 𝗺𝗲𝗺𝗼𝗿𝘆 𝗵𝗼𝗴𝘀. We're talking 10GB of memory for hours just to build an index for millions of vectors.
3️⃣ 𝗜𝗻𝘀𝗲𝗿𝘁𝗶𝗻𝗴 𝗻𝗲𝘄 𝘃𝗲𝗰𝘁𝗼𝗿𝘀 𝗶𝘀 𝗮 𝗽𝗮𝗶𝗻. With IVFFlat, you have to reindex preiodically. With HNSW, you're looking at graph traversals and memory allocation for every new vector.
4️⃣ 𝗠𝗲𝘁𝗮𝗱𝗮𝘁𝗮 𝗳𝗶𝗹𝘁𝗲𝗿𝗶𝗻𝗴 𝗶𝘀𝗻'𝘁 𝗰𝗹𝗲𝗮𝗿-𝗰𝘂𝘁. Pre-filtering is great for selective filters, but post-filtering can give you incomplete results, and it only gets worse with more filters.
5️⃣ 𝗡𝗼 𝗯𝘂𝗶𝗹𝘁-𝗶𝗻 𝗵𝘆𝗯𝗿𝗶𝗱 𝘀𝗲𝗮𝗿𝗰𝗵. You'll have to figure out how to balance vector similarity with text relevance and normalize scores on your own.

A new extension, pgvectorscale, does fix some of these problems with more memory-efficient indexing and better filter performance. But it isn't supported out-of-the-box on Azure and AWS, so you'd have to manage your own Postgres instance.

𝗽𝗴𝘃𝗲𝗰𝘁𝗼𝗿 is a great tool, but it's not a silver bullet. Make sure you're ready for the operational overhead before you commit.

🔗 Link to article: https://lnkd.in/d6jSRy2y


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

Tired of blocking I/O messing up your async Python apps?

Try 𝗮𝗶𝗼𝗳𝗶𝗹𝗲𝘀 - the game-changer for handling files in asyncio without the headaches.

It's an Apache-licensed library that wraps standard file ops into async magic.

Why you should use it:
• Mirrors Python's built-in file API, so it's super familiar.
• Supports buffered text, binary files, and even tempfiles asynchronously.
• Delegates to a thread pool to keep your event loop flying.


🔗 Link to repo: github(dot)com/Tinche/aiofiles

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

Vector databases are bloated.

What if you could get the same power with 97% less space?

Try 𝗟𝗘𝗔𝗡𝗡, the world's smallest vector index.

It transforms your laptop into a private and powerful RAG system.

You can finally RAG on everything.
-> Your entire file system
-> Millions of local emails
-> 💻 Your entire codebase
-> 💬 All your chat histories

Why is 𝗟𝗘𝗔𝗡𝗡 interesting?
✅ 𝗜𝘁'𝘀 𝗶𝗻𝗰𝗿𝗲𝗱𝗶𝗯𝗹𝘆 𝗹𝗶𝗴𝗵𝘁𝘄𝗲𝗶𝗴𝗵𝘁: Index 60 million documents in just 6GB instead of 201GB.
✅ 𝗜𝘁'𝘀 𝟭𝟬𝟬% 𝗽𝗿𝗶𝘃𝗮𝘁𝗲: Your data never leaves your laptop. No cloud costs, period.
✅ 𝗧𝗵𝗲𝗿𝗲 𝗶𝘀 𝗻𝗼 𝗮𝗰𝗰𝘂𝗿𝗮𝗰𝘆 𝗹𝗼𝘀𝘀: Get the same search quality as heavyweight solutions.

🔗 Link to repo: github(dot)com/yichuan-w/LEANN


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

Do you want to identify which lines use the most amount of memory in your Python script?

Try 𝗺𝗲𝗺𝗼𝗿𝘆_𝗽𝗿𝗼𝗳𝗶𝗹𝗲𝗿.

𝗺𝗲𝗺𝗼𝗿𝘆_𝗽𝗿𝗼𝗳𝗶𝗹𝗲𝗿 is a Python module to make a line-by-line analysis of memory consumption for Python functions.

Below you can see how to use memory_profiler within your Python script:

• Decorate the function you want to profile with @𝗽𝗿𝗼𝗳𝗶𝗹𝗲.
• Run the script by passing the option -𝗺 𝗺𝗲𝗺𝗼𝗿𝘆_𝗽𝗿𝗼𝗳𝗶𝗹𝗲𝗿 to load the 𝗺𝗲𝗺𝗼𝗿𝘆_𝗽𝗿𝗼𝗳𝗶𝗹𝗲𝗿 module.

Link to project: pypi(dot)org/project/memory-profiler


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

What, if an API call in your program fails?

Because of, let’s say, an instable internet connection?

This is not so uncommon.

You usually should have some retry mechanism in your program.

With 𝐭𝐞𝐧𝐚𝐜𝐢𝐭𝐲 in Python, this isn’t a problem anymore.

𝐭𝐞𝐧𝐚𝐜𝐢𝐭𝐲 offers a retrying behavior with a decorator with powerful features like:

- Define Stop Conditions
- Define Wait Conditions
- Customize retrying on Exception
- Retry on Coroutines

One of my favorite libraries!

Check it out: github(dot)com/jd/tenacity


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

OpenAI's Responses API isn't what they're telling you it is.

They are talking about how it's "faster, more flexible, easier" than the old chat completions API.

But here's what's really going on.

OpenAI hides their reasoning traces.

Unlike Claude, DeepSeek, or Qwen, you can't see GPT-5's chain of thought.

This creates a massive problem for developers using the stateless chat completions API.

Without access to reasoning traces, you can't pass them back into the conversation.

Result? GPT-5 performs worse than it should in your applications.

The Responses API is the workaround.

✅ It's stateful (OpenAI stores conversation history on their servers)
✅ They maintain the hidden reasoning traces in the backend
✅ They plug it into conversations automatically
✅ They strip it out before sending responses back to you

You get full model capability without ever seeing the reasoning.

Here's the uncomfortable truth:

The old stateless API is actually simpler.

Prefix caching works in both. Parallel tool calling works in both.

There's nothing inherently better about stateful architecture here.

The Responses API only exists because OpenAI won't show you what their models are thinking.

Meanwhile, Anthropic keeps their reasoning traces completely transparent.

If you're building with AI APIs in 2026, this matters.

The question is whether you're comfortable with your provider hiding critical information from you.

Link to article: https://lnkd.in/drZjD9Xj


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

AI makes your team faster.

But it also puts your data at risk.

Files are moving to new tools constantly.

AI models are training on things they shouldn't see.

And old security tools just can't see what is happening.

They are blind to the context that actually matters.

That is why I'm watching Cyberhaven’s launch on February 3.

They are revealing a new platform that combines data security and protection into one tool.

It's built to handle how messy and fast data moves today.

Join the live event to see:
📍 A live demo of this new AI-native platform.
📍 How to spot the few dangerous actions hidden inside millions of normal ones.
📍 Real stories from leaders on why old rules don't work anymore.

Don't wait for a data leak to find the holes in your security.

Grab your spot here: https://lnkd.in/dC-aGjjM

Struggling with slow data processing for AI training?

Try 𝗹𝗶𝘁𝗱𝗮𝘁𝗮, the Python library that transforms and optimizes datasets at scale!

𝗹𝗶𝘁𝗱𝗮𝘁𝗮 offers:
🟢 Streaming large cloud datasets without local downloads
🟢 Accelerating training by 20x with optimized data
🟢 Parallelizing tasks across multiple machines effortlessly
🟢 Securely handling sensitive data with advanced encryption

Perfect for images, text, video, and more.

Check it out: github(dot)com/Lightning-AI/litdata


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

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