I’ve been working in AI for close to a decade, starting well before today’s LLM-driven wave.

My work is grounded in understanding AI across layers — from hardware, AI models, and data to system design, reliability, and scale. I approach AI as an end-to-end engineering problem: not just modeling, but how models interact with data, infrastructure, and real-world constraints.

Alongside this, I write The AI Merge, a technical newsletter on AI foundations and systems engineering, focused on clear thinking and practical system design rather than hype.

I’m currently building an AI Systems course, centered on end-to-end system design, for engineers who want a deeper, systems-first understanding of modern AI.

🗞️ Read The AI Merge: https://read.theaimerge.com

Alex Razvant

Read more code.

And fewer articles, tutorials, or surface-level books.

If you’re already a Software Engineer or AI / ML Engineer, one of the fastest ways to level up right now is to: 

Study how big AI teams write and ship code - in public. 

Many top companies and major contributors to the AI space have been open-sourcing their internal frameworks, kernels, or infrastructure libraries. 

You might feel overwhelmed by the rapid pace at which AI is going.
But this is a massive learning opportunity if you know where to look.

Learning doesn’t just happen from papers and tutorials, especially if you already build software.

It happens when you:
- Read real production code
- Study PRs or commit history
- See how teams structure repos, APIs, configs, tests

Check what gets reviewed, what gets rejected, and why.
This kind of learning will upgrade your engineering intuition. 

Here are some excellent codebases to learn from.

✅ From DeepSeek:
1. FlashMLA  - A library of high-performance attention kernels.
Link: https://buff.ly/XbFnPA7 

2. DeepEP - A Communication Library for high-throughput MoE workloads.
Link: https://buff.ly/3qxMc3S

3. 3FS - A Distributed Filesystem for AI Workloads, for AI training and inference.
Link: https://buff.ly/ovfW47A

✅ From NVIDIA:
1. Dynamo - A datacenter-scale Inference Serving Framework for GenAI workloads.
Link: https://buff.ly/WyAMzBz 

2. NeMo  -  A Scalable Generative AI Framework, for training and deploying LLMs and VLMs.
Link: https://buff.ly/L9KTuJp 

✅ From vLLM:
1. The vLLM Library  - One of the most popular LLM inference engines, designed for scale.
Link: https://buff.ly/GLCfoIM

✅ From SGLang:
1. The SGLang Engine - A high-performance serving engine for LLMs & multimodal models.
Link: https://buff.ly/m4HhsWQ 

✅ From llama.cpp
1. The llama.cpp Engine - A minimal C/C++ inference engine for LLMs with broad hardware support and quantization options.
Link: https://buff.ly/u9yaXwp

📌 Don't study everything, you won't be able to.

Instead, pay attention to:
⇾ Code structure & abstractions
How are core functionalities separated (core logic vs hardware-specific optimizations)?

⇾  PRs & commit history
What context do authors provide? How do reviewers push back?

⇾  Interfaces & public APIs
How are public APIs designed?

⇾  Tests & benchmarks
How do the test suites look? What CI pipelines do they have, profiling? 

Don’t just use these libraries. 
Read them.

That will build your intuition and help you get better at your craft, either as a curious learner or an engineer as part of a larger team at your job.

---
I distill my learnings with my audience on The AI Merge.
Check it here (subscribe if interested) https://buff.ly/2rwqlNL

Senior AI Engineer | Writing The AI Merge Newsletter | Helping Engineers Build AI Beyond Demos

Continental

Enabled developers to solve issues in Cmake files to build C++ packages.
• Automatized Gtest, Doxygen, CTC, Traceability and QAC reports for repositories to achieve ASPICE requeriments.
• Launched Controllers in AWS with Jenkins Image to enable continous integration of repositories.
• Monitored controllers in Kibana dashboards.
• Offered solutions to teams regarding Continuous Testing.

Senior Continuous Integration Engineer Support

Verified and Integrated ARXML architecture in Tresos and AUTOSAR Builder to receive and send more signals in the IUC SW.
• Run CAPL Scripts for Integration Testing in Vector CANOE and debugging Software in Trace32.

Senior Software Architect and Integrator for Radar project

Led a team of 6 engineers through developing auto-alignment and extrinsic calibration LiDAR algorithms.
• Created Proof of Concept(PoC) Python scripts for auto-alignment and extrinsic calibration using computer vision packages as a OpenCV, PCL and Open3D and implemented perception algorithm such as RANSAC for Ground Detection, DBSCAN for Object Detection and Tracking and Gaussian filter for Line Detection.
• Developed C++ code implementation for Simulation-in-Loop environment based on eCAL-nodes and protobuffer standard communication.
• Enabled Gtest, Doxygen documentation, QAC, CTC reports for C++ Code using Cmake files also automatized using Jenkins and saved the artifacts in Artifactory.
• Implemented Key Performance Indicator (KPI) scripts in python and matlab and trigerred using Github Actions.
• Flashed code and real time debugging in ECU using Trace32.
• Implemented SAFe framework to coordinate the creation and increment of Value with the ART roadmap.
• Mantained backlog, prioritized tasks, converted from L2/L3 requirements to tickets, define Definition of Done and ready of tasks, performed Agile meetings, Analyzed metrics, created dashboard for sprint and planning in Jira.

Senior LiDAR Algorithm Engineer

Designed and implemented new algorithm for fast autocalibration in Advanced Radar Sensor (ARS510) project in order to achieve KPI Time for specific customer.

Senior Radar Algorithms Engineer

• Collaborated in the Point Cloud Algorithm Software component, creating a 3D point Cloud using raw sensor data.
• Fixes some code bugs raised by testing or customer in C++ and Kernel-C, showcasing espertise in debugging within simulation and hardware-in-loop enviroments.
• Created Detail desing of the component using doxygen comments.
• Implemated Dynamic Testing using Courage to complete the coverage of the code.
• Setup Cmake enviroment for building code.
• Analyzed raw sensor data with crosstalk and high/low Intensity scenarios. Implemented python notebooks using Jupyter for data analysis and Visualization in High Range Lidar(HRL) project.
• Acted as Scrum Master of the team, enabling colleagues in blockers and communicating with other teams, and also acted as Backup for Product Owner.

LiDAR Algorithm Engineer / Scrum Master

Designed and Developed Test Cases for different component for LiDAR sensor such as: Point Cloud Algorithm (PCA), Lidar Ego Motion (LEMO), Object Detection (OD) based on L3 requirements.
• Reviewed 100% of L3 Requirements for PCA component in IBM Doors, ensuring the coherency, linkage and atomicity.
• Developed 100% of Test Scripts from Scatch using python for validation of data and implemented KPI’s in PCA module.
• Used of High Performance Computing for Simulation-on-Loop and Jenkins for KPI’s performs.
• Collaborated and maintained contact with Functional delivery Owners and Algorithms Developers.
• Implemented Agile methodologic into test team.
• Extracted data from Oracle Database and transformed into dataframes and cleaned that data using Numpy and Pandas, to load it in MongoDB.

Entry|Intermediate Algo Test LiDAR Engineer

COVAF

Designed and Developed a Database of investment portfolios in SQL Server.
• Implemented the communication between SQL database and the cloud using Amazon Web Services such as EC2 for server, lambdas to create and call functions to run queries.

Jr Creative Developer

CIDESINL

Developed GUI for controlling and Data acquisition Motoman 6DOF robot and Force-Torque Sensor using Labview
• Implemented traveling salesman problem algorithm for robot route optimization in python.
• Design and Implemented a feed forward neural network for robot assembly tasks in python.

Robotics and Advanced Manufacturing Intern

Tliwaka Nao Team

Research Programmer

Mechatronics Engineer with 7+ years of experience in software programming and architecture, specializing in automotive systems and algorithms sensors for LiDAR and Radar for autonomous driving. Lastest roles were as a Senior  Software Architect for Advanced Radar Sensor and Short Range Radar Sensor, and also supporting Continuous Integration Platform at Continental. Machine Learning enthusiastic, passionate with  autonomous driving, artificial intelligence and computer vi sion; committed to create products and software solutions with quality’s standard in agile manner.

taplio

The Battle

Alex Razvant

Jonathan Ramiro Martínez Galindo

Followers

Posts

Engagements

Likes

Performance ScoreAlex Razvant's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Alex Razvant

Jonathan Ramiro Martínez Galindo

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);}Alex Razvant's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreAlex Razvant's LinkedIn Performance