Giacomo Veneri retweeted
Everyone is fine-tuning LLMs. Almost nobody understands what is actually being updated inside the model.
That distinction matters because LoRA, QLoRA, LoRA-FA, VeRA, Delta-LoRA and LoRA+ are usually discussed as if they were small variations of the same method. They are not. Some reduce trainable parameters, some reduce activation memory, some reduce the memory occupied by the frozen model, and some change the way the adapter itself is optimized.
I'm actually sharing 6 techniques.
1/ LoRA
Suppose a layer contains a pretrained weight matrix W. Full fine-tuning would update W directly. LoRA leaves W frozen and represents the update using two much smaller matrices, A and B, so that ΔW = BA.
For a square d × d weight matrix, full fine-tuning has d² parameters available to update. A rank-r LoRA adapter has roughly 2dr trainable parameters instead, where r is normally much smaller than d. This is the basic reason LoRA can adapt very large models without training every parameter in them.
2/ LoRA-FA
Standard LoRA trains both A and B. LoRA-FA freezes A and trains only B.
There is a useful reason for doing this beyond simply reducing the number of trainable parameters. Computing the gradient for A requires retaining the layer input activation. If A is fixed, that gradient is no longer required, which allows LoRA-FA to reduce activation memory as the LoRA rank grows.
3/ QLoRA
QLoRA attacks a different part of the memory problem. LoRA makes the adapter small, but the frozen base model can still occupy tens of gigabytes.
QLoRA keeps the base model frozen in 4-bit form and trains LoRA adapters through it. The original work used NF4, double quantization and paged optimizers, and demonstrated fine-tuning a 65B model on a single 48GB GPU.
This is an important distinction: QLoRA is not simply "LoRA with smaller adapters." The large memory saving comes from quantizing the frozen base model.
4/ VeRA
VeRA reduces the adapter itself further. Instead of learning a separate A and B for every adapted layer, it uses frozen random low-rank matrices that can be shared across layers, while learning much smaller scaling vectors.
The low-rank basis is therefore fixed. Training mainly determines how strongly different parts of that basis should contribute. This is why VeRA can use considerably fewer trainable parameters than ordinary LoRA.
5/ Delta-LoRA
Ordinary LoRA treats W as fixed throughout training. Delta-LoRA relaxes that constraint.
A and B are still trained, but the change in their product from one training step to the next is also used to update W. The base weights can therefore move without maintaining the ordinary gradients and optimizer states that full fine-tuning would require for W.
6/ LoRA+
LoRA+ does not introduce another adapter structure. It keeps W frozen and still trains A and B.
Its change is in the optimizer: A and B use different learning rates, with B receiving a larger rate. The motivation is that the two LoRA matrices do not behave identically during optimization, so forcing them to use the same learning rate is not necessarily the best choice.
Once these are separated by what they actually change, the family becomes much easier to understand.
LoRA reduces the number of weights being trained. LoRA-FA also targets activation memory. QLoRA compresses the frozen base model. VeRA reduces the learned adapter parameters further. Delta-LoRA allows the pretrained weights themselves to evolve through low-rank changes, while LoRA+ keeps the LoRA structure and changes its optimization.
That is really what PEFT is about => deciding which parts of a very large model actually need to move during adaptation, and which parts can remain fixed.
RaspBot and ROS2 integration with Agentic AI venergiac.substack.com/p/ras…
1
Giacomo Veneri retweeted
Andrej Karpathy’s 1-hour Stanford lecture on AI engineering is one of the best explanations I’ve seen of how AI systems actually work.
The progression is simple:
10% - LLM
30% - Prompt
50% - Agent
70% - Loop
100% - Graph
The key takeaway: AI engineering isn’t just about writing better prompts.
It’s about building systems around models, giving them context, memory, tools, feedback loops, and data flows.
“Delete everything, keep Graph.”
Definitely worth watching if you’re building with AI agents.
Watch and bookmark it.
From Predictive Analytics to Decision Governance: An AI-Enabled Framework for Condition-Based Maintenance of Industrial Gas Turbines onepetro.org/SPEADIP/proceed…
2
Giacomo Veneri retweeted
Memory is what allows Claude to learn from experience. In a new Engineering Blog post, we share how “dreaming” helps Claude organize what it’s learned, correct its mistakes, and find what it needs more easily the next time. anthropic.com/research/compo…
Giacomo Veneri retweeted
NVIDIA Open-Sourced Auto Ontology, a Semantic Layer for AI Agents!
Auto Ontology is a Python framework for building semantic, graph-based reasoning layers over data.
AI agents can already query databases and retrieve documents.
The harder problem is making sure they understand what the data actually means.
A metric like “revenue” might be defined differently across two systems. The same customer can appear under different identifiers. A generated SQL query can be syntactically correct and still use the wrong business definition.
Auto Ontology is built for that layer in between.
It can build a semantic catalog over enterprise metadata so agents can reason about entities, metrics, relationships, lineage, and where trusted definitions come from.
The workflow is roughly:
Enterprise data + metadata
→ Auto Ontology
→ Semantic catalog / knowledge graph
→ AI agent
→ Grounded query + provenance
It can generate ontologies from existing metadata, connect to semantic models you already have, reconcile inconsistent definitions across systems, and expose that shared context through MCP, CLI, or API.
For structured data, it can also translate natural-language questions into SQL while validating the query against the ontology and certified definitions before returning the result.
So instead of an agent simply asking:
“How much revenue did we generate last month?”
it can understand which revenue definition to use, which tables contain the right data, how those entities relate, and where the final answer came from.
Auto Ontology currently supports relational sources including Databricks, Postgres, Snowflake, and DuckDB.
It also exposes an MCP server, which means agents can query the same semantic layer directly.
For enterprise agents, retrieval alone is not enough.
They also need a shared understanding of what the underlying data actually means.
100% open source.
I’ve shared the GitHub repo in the comments!
NVIDIA Open-Sources Auto Ontology, a Semantic Layer for AI Agents
Your boss asks the AI agent: "What was revenue last quarter?"
It answers in 3 seconds. Clean number. Nice chart.
But it used the wrong "revenue" column. 𝗡𝗼𝗯𝗼𝗱𝘆 𝗻𝗼𝘁𝗶𝗰𝗲𝘀.
That's the problem NVIDIA's new open-source library, Auto Ontology, goes after. Its README puts it plainly:
"…𝘢 𝘳𝘦𝘷𝘦𝘯𝘶𝘦 𝘮𝘦𝘵𝘳𝘪𝘤 𝘢𝘯𝘴𝘸𝘦𝘳𝘦𝘥 𝘧𝘢𝘴𝘵𝘦𝘳 𝘪𝘴 𝘸𝘰𝘳𝘵𝘩 𝘯𝘰𝘵𝘩𝘪𝘯𝘨 𝘪𝘧 𝘵𝘩𝘦 𝘯𝘶𝘮𝘣𝘦𝘳 𝘪𝘴 𝘸𝘳𝘰𝘯𝘨."
𝗪𝗵𝗮𝘁 𝗶𝘁 𝗶𝘀:
A "context layer" that sits between an AI agent and your database. It tells the agent where the data lives, what each field means, and whether to use it.
𝗛𝗼𝘄 𝗶𝘁 𝘄𝗼𝗿𝗸𝘀, 𝘀𝘁𝗲𝗽 𝗯𝘆 𝘀𝘁𝗲𝗽:
• 𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿: it reads your database metadata and drafts an ontology, a map of business terms like "active customer"
• 𝗚𝗿𝗼𝘂𝗻𝗱: agents use those terms instead of guessing from raw table names
• 𝗩𝗮𝗹𝗶𝗱𝗮𝘁𝗲: it checks generated SQL against certified definitions
• 𝗘𝘅𝗽𝗼𝘀𝗲: agents reach it through MCP, a CLI or an API
• 𝗟𝗲𝗮𝗿𝗻: certified corrections go back into the map
𝗙𝗼𝗿 𝘂𝘀𝗲𝗿𝘀:
Ask a question, get 𝘁𝗵𝗲 𝗮𝗻𝘀𝘄𝗲𝗿, 𝘁𝗵𝗲 𝗦𝗤𝗟 𝗶𝘁 𝗿𝗮𝗻, 𝗮𝗻𝗱 𝘁𝗵𝗲 𝗿𝗼𝘄𝘀. You can check the work, not just trust it.
𝗙𝗼𝗿 𝗯𝘂𝗶𝗹𝗱𝗲𝗿𝘀:
• Apache 2.0, runs with Docker Compose or Helm
• Works with Databricks, Postgres, Snowflake and DuckDB
• Defaults to NVIDIA's Nemotron models, but 𝗮𝗻𝘆 𝗢𝗽𝗲𝗻𝗔𝗜-𝗰𝗼𝗺𝗽𝗮𝘁𝗶𝗯𝗹𝗲 𝗲𝗻𝗱𝗽𝗼𝗶𝗻𝘁 𝘄𝗼𝗿𝗸𝘀
• The MCP server has 𝟵 𝗿𝗲𝗮𝗱-𝗼𝗻𝗹𝘆 𝘁𝗼𝗼𝗹𝘀. There is deliberately no tool to browse raw tables, so agents must go through the business terms
𝗜𝘁 𝗯𝘂𝗶𝗹𝗱𝘀 𝗼𝗻 𝗲𝗮𝗿𝗹𝗶𝗲𝗿 𝘄𝗼𝗿𝗸:
• Cube and dbt MetricFlow: semantic layers you write by hand in code
• Wren AI: an open semantic layer for AI agents, with a context layer built partly by agents
• Snowflake's Semantic View Autopilot: also drafts semantic models with AI, but only for Snowflake
What's different here: 𝗼𝗽𝗲𝗻 𝘀𝗼𝘂𝗿𝗰𝗲, 𝘄𝗼𝗿𝗸𝘀 𝗮𝗰𝗿𝗼𝘀𝘀 𝗳𝗼𝘂𝗿 𝗱𝗮𝘁𝗮𝗯𝗮𝘀𝗲𝘀, 𝗮𝗻𝗱 𝗯𝘂𝗶𝗹𝘁 𝗳𝗼𝗿 𝗮𝗴𝗲𝗻𝘁𝘀 𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝗠𝗖𝗣.
𝗧𝗵𝗲 𝗵𝗼𝗻𝗲𝘀𝘁 𝗽𝗮𝗿𝘁:
• 𝗧𝗵𝗲 "𝟵𝟬% 𝗳𝗮𝘀𝘁𝗲𝗿" 𝗰𝗹𝗮𝗶𝗺 𝗶𝘀 𝗡𝗩𝗜𝗗𝗜𝗔'𝘀 𝗼𝘄𝗻 𝘁𝗲𝘀𝘁𝗶𝗻𝗴, with no baseline published
• 𝗡𝗼 𝗮𝗰𝗰𝘂𝗿𝗮𝗰𝘆 𝘀𝗰𝗼𝗿𝗲𝘀 𝗮𝗿𝗲 𝗽𝘂𝗯𝗹𝗶𝘀𝗵𝗲𝗱. NVIDIA ships an eval harness (BIRD, 1,534 questions) but no results
• That harness was tested with 𝗖𝗹𝗮𝘂𝗱𝗲 𝗢𝗽𝘂𝘀 𝟰.𝟴, 𝗻𝗼𝘁 𝗡𝗲𝗺𝗼𝘁𝗿𝗼𝗻
• 𝗪𝗶𝘁𝗵𝗼𝘂𝘁 𝗛𝗮𝘀𝗵𝗶𝗖𝗼𝗿𝗽 𝗩𝗮𝘂𝗹𝘁, 𝗱𝗮𝘁𝗮𝗯𝗮𝘀𝗲 𝗽𝗮𝘀𝘀𝘄𝗼𝗿𝗱𝘀 𝗮𝗿𝗲 𝘀𝘁𝗼𝗿𝗲𝗱 𝘂𝗻𝗲𝗻𝗰𝗿𝘆𝗽𝘁𝗲𝗱 in Postgres (per the README)
• It's young: 𝗮𝗯𝗼𝘂𝘁 𝟳𝟱𝟬 𝗚𝗶𝘁𝗛𝘂𝗯 𝘀𝘁𝗮𝗿𝘀, and NVIDIA isn't accepting outside contributions
𝗧𝗵𝗲 𝗯𝗶𝗴𝗴𝗲𝗿 𝗽𝗼𝗶𝗻𝘁: the hard part of AI analytics isn't writing SQL. 𝗜𝘁'𝘀 𝗮𝗴𝗿𝗲𝗲𝗶𝗻𝗴 𝗼𝗻 𝘄𝗵𝗮𝘁 𝘁𝗵𝗲 𝘄𝗼𝗿𝗱𝘀 𝗺𝗲𝗮𝗻.
GitHub: github.com/NVIDIA/auto-ontol…
Links in reply 👇
Recent academic and industrial application contributions mark the transition of Agentic AI from a purely theoretical or conceptual model to on-the-field implementation venergiac.substack.com/p/rec…
8
Hands-On Industrial Internet of Things: Build robust I-IoT infrastructur... youtu.be/mLqbW5jBUYk?si=x3kn… via @YouTube
7
I just published Do You Really Need an AI Strategy? medium.com/@venergiac/do-you…
7
I just published Fractal Chain of Thought (FCoT) for Physical Agentic AIoT medium.com/@venergiac/fracta…
1
10
Un Posto Vacante: Un papa scomparso. Un'antica reliquia riaffiorata. Un codice nascosto nel capolavoro più celebre del Rinascimento. GRATSI SU AMAZON OGGI amzn.eu/d/0gEeOj57 #Amazon via @Amazon
8
Un Posto Vacante: Un papa scomparso. Un'antica reliquia riaffiorata. Un codice nascosto nel capolavoro più celebre del Rinascimento. amzn.eu/d/0g0x5M1s
5
📢 Call for Papers | NDT Special Issue
Application of Artificial Intelligence and Machine Learning Methods to NDT Signal and Image Analysis
📅 Submission deadline: 3 Apr 2027
🔗 brnw.ch/21x5dvW
#NDT #ArtificialIntelligence #MachineLearning #DeepLearning #SHM
Application of Artificial Intelligence and Machine Learning Methods to NDT Signal and Image Analysis mdpi.com/si/288125 #mdpindt via @NDT_MDPI
1
9
Giacomo Veneri retweeted
EL INGENIERO QUE CREÓ CLAUDE CODE ACABA DE SUBIR 28 MINUTOS DE ORO PURO
sacó un video completo sobre cómo escribir prompts que realmente entregan resultados, no genéricos de manual.
hay cursos de 300 dólares que no llegan ni a la suela de lo que explica en los primeros 10 minutos.
archivos CLAUDE.md, atajos de memoria, sesiones paralelas, patrones de prompting que casi nadie usa.
28 minutos. GRATIS. sin relleno.
no importa si llevas años programando o recién empiezas con Claude: esto te cambia el flujo de trabajo desde hoy.
la documentación oficial te dice qué existe. esto te enseña cómo usarlo de verdad.
guárdalo. míralo. compártelo.
Giacomo Veneri: The Most Visionary Digital Engineering Leader Driving Physical AI & Industrial IoT Innovation in 2026 theempiremagazine.com/magazi…
2
Agentic AIOT : The Future of Manufacturing: Agentic AI, Industrial IoT, and Predictive Maintenance open.substack.com/pub/venerg…
10
The Big “AI Strategy” Bluff: What If We Are All Chasing a Phantom? open.substack.com/pub/venerg…
2














