| Server IP : 3.147.158.171 / Your IP : 216.73.216.88 Web Server : Apache/2.4.67 (Amazon Linux) OpenSSL/3.5.5 System : Linux ip-172-31-2-178.us-east-2.compute.internal 6.1.172-216.329.amzn2023.x86_64 #1 SMP PREEMPT_DYNAMIC Wed May 20 06:31:34 UTC 2026 x86_64 User : ec2-user ( 1000) PHP Version : 8.4.21 Disable Function : NONE MySQL : OFF | cURL : ON | WGET : ON | Perl : ON | Python : OFF | Sudo : ON | Pkexec : OFF Directory : /tsai/repo/api/logs/articles/ |
Upload File : |
# Tavily Search Results: Geneva
**Query:** latest updates agentic AI coding
**Topic:** general
**Results:** 8
**Date:** 2026-01-01 07:14:27
---
## 1. AI Coding Tools in 2025: Welcome to the Agentic CLI Era
**URL:** [https://thenewstack.io/ai-coding-tools-in-2025-welcome-to-the-agentic-cli-era/](https://thenewstack.io/ai-coding-tools-in-2025-welcome-to-the-agentic-cli-era/)
**Score:** 0.9998
### Summary
On re-reading the article now, I quoted the term “agentic” only once and never use the term CLI. It wasn’t until the end of May — when Claude Opus, a stronger Large Language Model (LLM), was introduced — that I mentioned the term “agentic” multiple times. [...] From a software development perspective, there is little doubt what 2025 will be known for when looking back. In fact, it already has a somewhat wild name: the Agentic Era. It didn’t even need a full year to pick up that moniker. [...] Nov 20th 2025 11:00am, by Cynthia Dunlop
AI AI Engineering API Management Backend development Data Frontend Development Large Language Models Security Software Development WebAssembly
Google's New Gemini 3 Flash Rivals Frontier Models at a Fraction of the Cost
Dec 17th 2025 8:00am, by Frederic Lardinois
AI Coding Tools in 2025: Welcome to the Agentic CLI Era
Dec 17th 2025 6:00am, by David Eastman
A Frontier Model Built Like a Brain with Python and Rust
### Full Content
AI / [AI Agents](https://thenewstack.io/ai-agents/) / [Developer tools](https://thenewstack.io/developer-tools/) / [Software Development](https://thenewstack.io/software-development/)" name="x-tns-categories">David Eastman" name="x-tns-authors"> AI Coding Tools in 2025: Welcome to the Agentic CLI Era - The New Stack
TNS
OK
SUBSCRIBE
Join our community of software engineering leaders and aspirational developers. Always stay in-the-know by getting the most important news and exclusive content delivered fresh to your inbox to learn more about at-scale software development.
EMAIL ADDRESS
REQUIRED
SUBSCRIBE
RESUBSCRIPTION REQUIRED
It seems that you've previously unsubscribed from our newsletter in the past. Click the button below to open the re-subscribe form in a new tab. When you're done, simply close that tab and continue with this form to complete your subscription.
RE-SUBSCRIBE
The New Stack does not sell your information or share it with unaffiliated third parties. By continuing, you agree to our [Terms of Use](/terms-of-use/) and [Privacy Policy](/privacy-policy/).
Welcome and thank you for joining The New Stack community!
Please answer a few simple questions to help us deliver the news and resources you are interested in.
FIRST NAME
REQUIRED
LAST NAME
REQUIRED
COMPANY NAME
REQUIRED
COUNTRY
REQUIRED
Select ... United States Canada India United Kingdom Germany France --- Afghanistan Albania Algeria American Samoa Andorra Angola Anguilla Antarctica Antigua and Barbuda Argentina Armenia Aruba Asia/Pacific Region Australia Austria Azerbaijan Bahamas Bahrain Bangladesh Barbados Belarus Belgium Belize Benin Bermuda Bhutan Bolivia Bonaire, Sint Eustatius and Saba Bosnia and Herzegovina Botswana Bouvet Island Brazil British Indian Ocean Territory Brunei Darussalam Bulgaria Burkina Faso Burundi Cambodia Cameroon Canada Cape Verde Cayman Islands Central African Republic Chad Chile China Christmas Island Cocos (Keeling) Islands Colombia Comoros Congo Congo, The Democratic Republic of the Cook Islands Costa Rica Croatia Cuba Curaçao Cyprus Czech Republic Côte d'Ivoire Denmark Djibouti Dominica Dominican Republic Ecuador Egypt El Salvador Equatorial Guinea Eritrea Estonia Ethiopia Falkland Islands (Malvinas) Faroe Islands Fiji Finland France French Guiana French Polynesia French Southern Territories Gabon Gambia Georgia Germany Ghana Gibraltar Greece Greenland Grenada Guadeloupe Guam Guatemala Guernsey Guinea Guinea-Bissau Guyana Haiti Heard Island and Mcdonald Islands Holy See (Vatican City State) Honduras Hong Kong Hungary Iceland India Indonesia Iran, Islamic Republic Of Iraq Ireland Isle of Man Israel Italy Jamaica Japan Jersey Jordan Kazakhstan Kenya Kiribati Korea, Republic of Kuwait Kyrgyzstan Laos Latvia Lebanon Lesotho Liberia Libyan Arab Jamahiriya Liechtenstein Lithuania Luxembourg Macao Madagascar Malawi Malaysia Maldives Mali Malta Marshall Islands Martinique Mauritania Mauritius Mayotte Mexico Micronesia, Federated States of Moldova, Republic of Monaco Mongolia Montenegro Montserrat Morocco Mozambique Myanmar Namibia Nauru Nepal Netherlands Netherlands Antilles New Caledonia New Zealand Nicaragua Niger Nigeria Niue Norfolk Island North Korea North Macedonia Northern Mariana Islands Norway Oman Pakistan Palau Palestinian Territory, Occupied Panama Papua New Guinea Paraguay Peru Philippines Pitcairn Islands Poland Portugal Puerto Rico Qatar Reunion Romania Russian Federation Rwanda Saint Barthélemy Saint Helena Saint Kitts and Nevis Saint Lucia Saint Martin Saint Martin Saint Pierre and Miquelon Saint Vincent and the Grenadines Samoa San Marino Sao Tome and Principe Saudi Arabia Senegal Serbia Serbia and Montenegro Seychelles Sierra Leone Singapore Sint Maarten Slovakia Slovenia Solomon Islands Somalia South Africa South Georgia and the South Sandwich Islands South Sudan Spain Sri Lanka Sudan Suriname Svalbard and Jan Mayen Swaziland Sweden Switzerland Syrian Arab Republic Taiwan Tajikistan Tanzania, United Republic of Thailand Timor-Leste Togo Tokelau Tonga Trinidad and Tobago Tunisia Turkey Turkmenistan Turks and Caicos Islands Tuvalu Uganda Ukraine United Arab Emirates United Kingdom United States United States Minor Outlying Islands Uruguay Uzbekistan Vanuatu Venezuela Vietnam Virgin Islands, British Virgin Islands, U.S. Wallis and Futuna Western Sahara Yemen Zambia Zimbabwe Åland Islands
ZIPCODE
REQUIRED
Great to meet you!
Tell us a bit about your job so we can cover the topics you find most relevant.
What is your job level?
REQUIRED
Select ... C-Level VP/Director Manager/Supervisor Mid Level or Senior Non-Managerial Staff Entry Level/Junior Staff Freelancer/Contractor Student/Intern Other ...
Which of these most closely describes your job role?
REQUIRED
Select ... Developer/Software Engineer SysAdmin/Operations/SRE Architect Security Professional DevOps Engineer/Team Community Manager/Develo
*[Content truncated...]*
---
## 2. Agentic ProbLLMs: Exploiting AI Computer-Use and Coding Agents
**URL:** [https://media.ccc.de/v/39c3-agentic-probllms-exploiting-ai-computer-use-and-coding-agents](https://media.ccc.de/v/39c3-agentic-probllms-exploiting-ai-computer-use-and-coding-agents)
**Score:** 0.9997
### Summary
During the Month of AI Bugs (August 2025), I responsibly disclosed over two dozen security vulnerabilities across all major agentic AI coding assistants. This talk distills the most severe findings and patterns observed.
Key highlights include:
Critical prompt-injection exploits enabling zero-click data exfiltration and arbitrary remote code execution across multiple platforms and vendor products [...] This talk demonstrates end-to-end prompt injection exploits that compromise agentic systems. Specifically, we will discuss exploits that target computer-use and coding agents, such as Anthropic's Claude Code, GitHub Copilot, Google Jules, Devin AI, ChatGPT Operator, Amazon Q, AWS Kiro, and others.
### Full Content
Agentic ProbLLMs: Exploiting AI Computer-Use and Coding Agents - media.ccc.de
===============
[](https://media.ccc.de/)
[](https://media.ccc.de/about.html)
* [System](https://media.ccc.de/v/39c3-agentic-probllms-exploiting-ai-computer-use-and-coding-agents#)
* [Light](https://media.ccc.de/v/39c3-agentic-probllms-exploiting-ai-computer-use-and-coding-agents#)
* [Dark](https://media.ccc.de/v/39c3-agentic-probllms-exploiting-ai-computer-use-and-coding-agents#)
[News](https://media.ccc.de/news.atom)
[RSS, last 100](https://media.ccc.de/updates.rdf)
[Podcast feed of the last two years](https://media.ccc.de/podcast-hq.xml)[SD quality](https://media.ccc.de/podcast-lq.xml "Podcast feed of the last two years (SD)")
[Podcast audio feed of the last year](https://media.ccc.de/podcast-audio-only.xml)
[Podcast archive feed, everything older than two years](https://media.ccc.de/podcast-archive-hq.xml)[SD quality](https://media.ccc.de/podcast-archive-lq.xml "Podcast archive feed, everything older than two years (SD)")
Podcast feeds for 39c3
[opus](https://media.ccc.de/c/39c3/podcast/opus.xml "opus")
[webm](https://media.ccc.de/c/39c3/podcast/webm-hq.xml "webm")[SD quality](https://media.ccc.de/c/39c3/podcast/webm-lq.xml "webm (SD)")
[mp3](https://media.ccc.de/c/39c3/podcast/mp3.xml "mp3")
[mp4](https://media.ccc.de/c/39c3/podcast/mp4-hq.xml "mp4")[SD quality](https://media.ccc.de/c/39c3/podcast/mp4-lq.xml "mp4 (SD)")
[vtt](https://media.ccc.de/c/39c3/podcast/vtt.xml "vtt")
[News](https://media.ccc.de/news.atom)
[RSS, last 100](https://media.ccc.de/updates.rdf)
[Podcast feed of the last two years](https://media.ccc.de/podcast-hq.xml)[SD quality](https://media.ccc.de/podcast-lq.xml "Podcast feed of the last two years (SD)")
[Podcast audio feed of the last year](https://media.ccc.de/podcast-audio-only.xml)
[Podcast archive feed, everything older than two years](https://media.ccc.de/podcast-archive-hq.xml)[SD quality](https://media.ccc.de/podcast-archive-lq.xml "Podcast archive feed, everything older than two years (SD)")
Podcast feeds for 39c3
[opus](https://media.ccc.de/c/39c3/podcast/opus.xml "opus")
[webm](https://media.ccc.de/c/39c3/podcast/webm-hq.xml "webm")[SD quality](https://media.ccc.de/c/39c3/podcast/webm-lq.xml "webm (SD)")
[mp3](https://media.ccc.de/c/39c3/podcast/mp3.xml "mp3")
[mp4](https://media.ccc.de/c/39c3/podcast/mp4-hq.xml "mp4")[SD quality](https://media.ccc.de/c/39c3/podcast/mp4-lq.xml "mp4 (SD)")
[vtt](https://media.ccc.de/c/39c3/podcast/vtt.xml "vtt")
1. [browse](https://media.ccc.de/b)
2. [congress](https://media.ccc.de/b/congress)
3. [2025](https://media.ccc.de/b/congress/2025)
4. event
[](https://media.ccc.de/c/39c3)
Agentic ProbLLMs: Exploiting AI Computer-Use and Coding Agents
==============================================================
[Johann Rehberger](https://media.ccc.de/search?p=Johann+Rehberger)
Video Player
[Close](https://media.ccc.de/v/39c3-agentic-probllms-exploiting-ai-computer-use-and-coding-agents#)
Media error: Format(s) not supported or source(s) not found
[Download File: https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-eng-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents.mp4](https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-eng-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents.mp4)[Download File: https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-fra-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents.mp4](https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-fra-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents.mp4)[Download File: https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-deu-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents.mp4](https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-deu-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents.mp4)[Download File: https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-eng-deu-fra-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents_hd.mp4](https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/h264-hd/39c3-1306-eng-deu-fra-Agentic_ProbLLMs_Exploiting_AI_Computer-Use_and_Coding_Agents_hd.mp4)[Download File: https://us.mirror.ionos.com/projects/media.ccc.de//congress/2025/webm-h
*[Content truncated...]*
---
## 3. Agentic AI from First Principles: Reflection | by Mariya Mansurova
**URL:** [https://miptgirl.medium.com/agentic-ai-from-first-principles-reflection-65e51cebf676](https://miptgirl.medium.com/agentic-ai-from-first-principles-reflection-65e51cebf676)
**Score:** 0.9995
### Summary
In “Reflexion: Language Agents with Verbal Reinforcement Learning” Shinn et al. (2023), the authors achieved a 91% pass@1 accuracy on the HumanEval coding benchmark, surpassing the previous state-of-the-art GPT-4, which scored just 80%. They also found that Reflexion significantly outperforms all baseline approaches on the HotPotQA benchmark (a Wikipedia-based Q&A dataset that challenges agents to parse content and reason over multiple supporting documents). [...] It’s time to wrap things up. In this article, we started our journey into understanding how the magic of agentic AI systems works. To figure it out, we’ll implement a multi-agent text-to-data tool using only API calls to foundation models. Along the way, we’ll walk through the key design patterns step by step: starting today with reflection, and moving on to tool use, planning, and multi-agent coordination. [...] To gain a deep understanding, we’ll build a multi-AI agent system from scratch. We’ll avoid using frameworks like CrewAI or smolagents and instead work directly with the foundation model API. Along the way, we’ll explore the fundamental agentic design patterns: reflection, tool use, planning, and multi-agent setups. Then, we’ll combine all this knowledge to build a multi-AI agent system that can answer complex data-related questions.
### Full Content
[Sitemap](/sitemap/sitemap.xml)
[Open in app](https://play.google.com/store/apps/details?id=com.medium.reader&referrer=utm_source%3DmobileNavBar&source=post_page---top_nav_layout_nav-----------------------------------------)
[Sign in](https://medium.com/m/signin?operation=login&redirect=https%3A%2F%2Fmiptgirl.medium.com%2Fagentic-ai-from-first-principles-reflection-65e51cebf676&source=post_page---top_nav_layout_nav-----------------------global_nav------------------)
[Sign in](https://medium.com/m/signin?operation=login&redirect=https%3A%2F%2Fmiptgirl.medium.com%2Fagentic-ai-from-first-principles-reflection-65e51cebf676&source=post_page---top_nav_layout_nav-----------------------global_nav------------------)
# Agentic AI from First Principles: Reflection
## From theory to code: building feedback loops that **improve LLM accuracy**
[Mariya Mansurova](/?source=post_page---byline--65e51cebf676---------------------------------------)
20 min read
·
6 days ago
--
Arthur C. Clarke’s Third Law says that “*any sufficiently advanced technology is indistinguishable from magic*”. That’s exactly how a lot of today’s AI frameworks feel. Tools like GitHub Copilot, Claude Desktop, OpenAI Operator, and Perplexity Comet are automating everyday tasks that would’ve seemed impossible to automate just five years ago. What’s even more remarkable is that with just a few lines of code, we can build our own sophisticated AI tools: ones that search through files, browse the web, click links, and even make purchases. It really does feel like magic.
Even though I [genuinely believe in data wizards](https://towardsdatascience.com/i-think-of-analysts-as-data-wizards-who-help-their-product-teams-solve-problems/), I don’t believe in magic. I find it exciting (and often helpful) to understand how things are actually built and what’s happening under the hood. That’s why I’ve decided to share a series of posts on agentic AI design concepts that’ll help you understand how all these magical tools actually work.
To gain a deep understanding, we’ll build a multi-AI agent system from scratch. We’ll avoid using frameworks like CrewAI or smolagents and instead work directly with the foundation model API. Along the way, we’ll explore the fundamental agentic design patterns: reflection, tool use, planning, and multi-agent setups. Then, we’ll combine all this knowledge to build a multi-AI agent system that can answer complex data-related questions.
As Richard Feynman put it, “*What I cannot create, I do not understand*.” So let’s start building! In this article, we’ll focus on the reflection design pattern. But first, let’s figure out what exactly reflection is.
## What reflection is
Let’s reflect on how we (humans) usually work on tasks. Imagine I need to share the results of a recent feature launch with my PM. I’ll likely put together a quick draft and then read it once or twice from beginning to end, ensuring that all parts are consistent, there’s enough information, and there are no typos.
Or let’s take another example: writing a SQL query. I’ll either write it step by step, checking the intermediate results along the way, or (if it’s simple enough) I’ll draft it all at once, execute it, look at the result (checking for errors or whether the result matches my expectations), and then tweak the query based on that feedback. I might rerun it, check the result, and iterate until it’s right.
So we rarely write long texts from top to bottom in one go. We usually circle back, review, and tweak as we go. These feedback loops are what help us improve the quality of our work.
LLMs use a different approach. If you ask an LLM a question, by default, it will generate an answer token by token, and the LLM won’t be able to review its result and fix any issues. But in an agentic AI setup, we can create feedback loops for LLMs too, either by asking the LLM to review and improve its own answer or by sharing external feedback with it (like the results of a SQL execution). And that’s the whole point of reflection. It sounds pretty straightforward, but it can yield significantly better results.
There’s a substantial body of research showing the benefits of reflection:
* **“**[**Self-Refine: Iterative Refinement with Self-Feedback**](https://arxiv.org/abs/2303.17651?utm_campaign=The+Batch&utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9dHVnW1I1bA3sPBbsikjT165Qez3QiiAssknCERwgki818YHG7PyHOQSgg-nxKDa0BuE7B)**”** Madaan et al. (2023) showed that self-refinement improved performance by ~20% across diverse tasks, ranging from dialogue response generation to mathematical reasoning.
* In **“**[**Reflexion: Language Agents with Verbal Reinforcement Learning**](https://arxiv.org/abs/2303.11366?utm_campaign=The+Batch&utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9dHVnW1I1bA3sPBbsikjT165Qez3QiiAssknCERwgki818YHG7PyHOQSgg-nxKDa0BuE7B)**”** Shinn et al. (2023), the authors achieved a 91% pass@1 accuracy on the HumanEval coding benchmark, surpassing th
*[Content truncated...]*
---
## 4. FDA Expands Artificial Intelligence Capabilities with ...
**URL:** [https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment)
**Score:** 0.9990
### Summary
Today’s agentic AI deployment will enable FDA staff to further advance the use of AI to assist with more complex tasks, such as meeting management, pre-market reviews, review validation, post-market surveillance, inspections and compliance and administrative functions.
As part of the agentic AI deployment, the agency is launching a two-month Agentic AI Challenge for staff to build Agentic AI solutions and demonstrate them at the FDA Scientific Computing Day in January 2026. [...] For Immediate Release:December 01, 2025
The U.S. Food and Drug Administration today announced the deployment of agentic AI capabilities for all agency employees. Agentic AI capabilities will enable the creation of more complex AI workflows — harnessing various AI models — to assist with multi-step tasks.
### Full Content
FDA Expands Artificial Intelligence Capabilities with Agentic AI Deployment | FDA
===============
* [Skip to main content](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment#main-content)
* [Skip to FDA Search](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment#search-form)
* [Skip to in this section menu](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment#section-nav)
* [Skip to footer links](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment#footer)
An official website of the United States government[Here’s how you know](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment#)
**The .gov means it’s official.**
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.
**The site is secure.**
The **https://** ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.
[](https://www.fda.gov/ "FDA Homepage")
* Search
* Menu
Search FDA Submit search
Featured
--------
* [Report a Product Problem](https://www.safetyreporting.hhs.gov/smarthub#/ "Report a Problem with an FDA Regulated Product")
* [Contact FDA](https://www.fda.gov/about-fda/contact-fda)
* [FDA Guidance Documents](https://www.fda.gov/regulatory-information/search-fda-guidance-documents)
* [Recalls, Market Withdrawals and Safety Alerts](https://www.fda.gov/safety/recalls-market-withdrawals-safety-alerts)
* [Press Announcements](https://www.fda.gov/news-events/newsroom/press-announcements)
* [Warning Letters](https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/compliance-actions-and-activities/warning-letters)
* [Advisory Committees](https://www.fda.gov/advisory-committees)
* [En Español](https://www.fda.gov/fda-en-espanol)
Products
--------
* [Food](https://www.fda.gov/food)
* [Drugs](https://www.fda.gov/drugs)
* [Medical Devices](https://www.fda.gov/medical-devices)
* [Radiation-Emitting Products](https://www.fda.gov/radiation-emitting-products)
* [Vaccines, Blood, and Biologics](https://www.fda.gov/vaccines-blood-biologics)
* [Animal and Veterinary](https://www.fda.gov/animal-veterinary)
* [Cosmetics](https://www.fda.gov/cosmetics)
* [Tobacco Products](https://www.fda.gov/tobacco-products)
Topics
------
* [About FDA](https://www.fda.gov/about-fda)
* [Combination Products](https://www.fda.gov/combination-products)
* [Regulatory Information](https://www.fda.gov/regulatory-information)
* [Safety](https://www.fda.gov/safety)
* [Emergency Preparedness](https://www.fda.gov/emergency-preparedness-and-response)
* [International Programs](https://www.fda.gov/international-programs)
* [News and Events](https://www.fda.gov/news-events)
* [Training and Continuing Education](https://www.fda.gov/training-and-continuing-education)
* [Inspections and Compliance](https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations)
* [Science and Research](https://www.fda.gov/science-research)
Information For
---------------
* [Consumers](https://www.fda.gov/consumers)
* [Patients](https://www.fda.gov/patients)
* [Industry](https://www.fda.gov/industry)
* [Health Professionals](https://www.fda.gov/health-professionals)
* [Federal, State and Local Officials](https://www.fda.gov/federal-state-local-tribal-and-territorial-officials)
[In this section: Press Announcements](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment#section-nav)
* [Press Announcements](https://www.fda.gov/news-events/fda-newsroom/press-announcements "Press Announcements")
1. [Home](https://www.fda.gov/)
2. [News & Events](https://www.fda.gov/news-events)
3. [FDA Newsroom](https://www.fda.gov/news-events/fda-newsroom)
4. [Press Announcements](https://www.fda.gov/news-events/fda-newsroom/press-announcements)
5. [FDA Expands Artificial Intelligence Capabilities with Agentic AI Deployment](https://www.fda.gov/news-events/press-announcements/fda-expands-artificial-intelligence-capabilities-agentic-ai-deployment)
1. [Press Announcements](https://www.fda.gov/news-events/fda-newsroom/p
*[Content truncated...]*
---
## 5. Practical Techniques and Tips | YK Sugi, AI By the Bay 2025
**URL:** [https://www.youtube.com/watch?v=m1HOq-T31cw](https://www.youtube.com/watch?v=m1HOq-T31cw)
**Score:** 0.9985
### Summary
Agentic Coding with Discipline and Skill: Practical Techniques and Tips. This talk distills hard-won lessons from spending hundreds of millions of tokens building production applications with Claude Code and other agentic coding tools. I'll demonstrate five critical techniques that have emerged from real-world practice: using agents for technical research when starting from zero knowledge, breaking complex features into agent-digestible tasks, letting agents complete their own build-test-fix [...] cycles, managing context effectively across long conversations, and collaborative debugging. Beyond these core techniques, I'll share how agentic coding has fundamentally changed my development practice - making me braver when diving into unfamiliar codebases and enabling rapid creation of one-off scripts and tools that would have been impractical to build manually. The new 10,000 hour rule in the world of agentic coding is the "billion token rule". Let's get started with the first few [...] Agentic Coding with Discipline & Skill: Practical Techniques and Tips | YK Sugi, AI By the Bay 2025 - YouTube
Back Image 1
Skip navigation
Search
Search with your voice
- [x] Include playlist
An error occurred while retrieving sharing information. Please try again later.
Image 6
0:00
")
0:00 / 27:08
Live
•Watch full video
•
•
### Full Content
Agentic Coding with Discipline & Skill: Practical Techniques and Tips | YK Sugi, AI By the Bay 2025 - YouTube
===============
Back [](https://www.youtube.com/ "YouTube Home")
Skip navigation
Search
Search with your voice
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
[Sign in](https://accounts.google.com/ServiceLogin?service=youtube&uilel=3&passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Faction_handle_signin%3Dtrue%26app%3Ddesktop%26hl%3Den%26next%3Dhttps%253A%252F%252Fwww.youtube.com%252Fwatch%253Fv%253Dm1HOq-T31cw&hl=en&ec=65620)
[](https://www.youtube.com/ "YouTube Home")
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
[Agentic Coding with Discipline & Skill: Practical Techniques and Tips | YK Sugi, AI By the Bay 2025](https://www.youtube.com/watch?v=m1HOq-T31cw)
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
Tap to unmute
2x
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
Agentic Coding with Discipline & Skill: Practical Techniques and Tips | YK Sugi, AI By the Bay 2025
---------------------------------------------------------------------------------------------------
FunctionalTV 26 views 3 weeks ago
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
Search
Copy link
Info
Shopping
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
If playback doesn't begin shortly, try restarting your device.
•
You're signed out
Videos you watch may be added to the TV's watch history and influence TV recommendations. To avoid this, cancel and sign in to YouTube on your computer.
Cancel Confirm
Video unavailable
[](https://www.youtube.com/watch?v=m1HOq-T31cw)
Share
[](https://www.youtube.com/watch?v=m1HOq-T31cw "Share link")- [x] Include playlist
An error occurred while retrieving sharing information. Please try again later.
0:00
[](https://www.youtube.com/watch?v=m1HOq-T31cw)[](https://www.youtube.com/watch?v=pJ0iJFmQG_w "Next (SHIFT+n)")
0:00 / 27:08
Live
•Watch full video
•
•
[27:45 Scaling Multimodal Data and Reinforcement Learning with Ray | Jaikumar Ganesh, AI By the Bay 2025 FunctionalTV 80 views • 3 weeks ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=pJ0iJFmQG_w)[36:31 Spec Driven Development: AI Assisted Coding With Doc at the Core | Bruno Borges, AI By the Bay 25 FunctionalTV 143 views • 9 days ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=g2JMV4JWRDs)[26:24 6 Months of Claude Code Lessons in 27 Minutes AI with Avthar 226K views • 3 months ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=rfDvkSkelhg)[25:57 Stop Rambling: The 3-2-1 Speaking Trick That Makes You Sound Like A CEO BigDeal by Codie Sanchez 432K views • 9 days ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=t260757b_vU)[12:35 How To Learn So Fast It’s Almost Unfair theMITmonk 970K views • 12 days ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=npQ2IORdlvU)[18:14 Ex-OpenAI Scientist WARNS: "You Have No Idea What's Coming"AI Upload 4.8M views • 5 months ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=79-bApI3GIU)[31:50 Can We Trust AI-Generated Code? | Baruch Sadogursky & Leonid Igolnik, AI By the Bay25 FunctionalTV 43 views • 9 days ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=vXdT-KGxz-E)[15:36 AI Coding Sucks Syntax and Coding Garden 231K views • 2 months ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=0ZUkQF6boNg)[35:44 Anthropic C.E.O.: Massive A.I. Spending Could Haunt Some Companies The New York Times and New York Times Events 260K views • Streamed 4 weeks ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=FEj7wAjwQIk)[22:47 THE most advanced city in the world?! How? Let me show you!Heideexyz 1.2M views • 3 weeks ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=SDRtxwM4Wrg)[32:32 The Strange Math That Predicts (Almost) Anything Veritasium 10M views • 5 months ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=KZeIEiBrT_w)[11:22 I Tried Google's 7 FREE AI Tools (Destroys All Paid Alternatives)Parker Prompts 154K views • 2 days ago Live Playlist ()Mix (50+)](https://www.youtube.com/watch?v=gXm4Rp6gjEU)
Agentic Coding with Discipline & Skill: Practical Techniques and Tips | YK Sugi, AI By the Bay 2025
===================================================================================================
[](https://www.youtube.com/@F
*[Content truncated...]*
---
## 6. From Chatbot to Code Threat: OWASP's Agentic AI Top 10 ...
**URL:** [https://www.legitsecurity.com/blog/from-chatbot-to-code-threat-owasps-agentic-ai-top-10-and-the-specialized-risks-of-coding-agents](https://www.legitsecurity.com/blog/from-chatbot-to-code-threat-owasps-agentic-ai-top-10-and-the-specialized-risks-of-coding-agents)
**Score:** 0.9980
### Summary
AI coding agents, including monitoring usage, managing technology-approval workflows, and enforcing policies directly within the IDE. [...] GenAI Code Untested by Security Controls:Allowing unvalidated agent-generated code into production creates cascading weaknesses(Relates to OWASP’s Cascading Failures risk). [...] As autonomous agents move from experimentation into real engineering workflows, their potential to introduce code-level vulnerabilities grows exponentially. The combined insights from OWASP and Legit’s research make one thing clear: securing AI Coding Agents isn’t optional-it’s now a foundational requirement for protecting your software supply chain.
Introducing VibeGuard: Secure Your AI Code Pipeline
### Full Content
From Chatbot to Code Threat: OWASP’s Agentic AI Top 10 and the Specialized Risks of Coding Agents
===============
[×](javascript:void(0);)
This website stores cookies on your computer. These cookies are used to improve your website experience and provide more personalized services to you, both on this website and through other media. To find out more about the cookies we use, see our Privacy Policy.
[Accept](javascript:void(0);)
[**Upcoming Webinars on January 20th: See more here!**](https://www.legitsecurity.com/events)
[](https://www.legitsecurity.com/)
* ASPM Platform ASPM Platform
Overview
Legit is an AI-native ASPM platform that automates AppSec issue discovery, prioritization, and remediation. A trusted ASPM vendor for your AppSec and software supply chain security programs.
[Learn More](https://www.legitsecurity.com/platform/aspm)[Integrations](https://www.legitsecurity.com/integrations) Use Cases: [Unified Vulnerability Remediation](https://www.legitsecurity.com/vulnerability-management) [](https://www.legitsecurity.com/vulnerability-management) [Code Security (SAST, SCA)](https://www.legitsecurity.com/press-releases/legit-security-launches-advanced-sca-and-sast-enhancements) [](https://www.legitsecurity.com/press-releases/legit-security-launches-advanced-sca-and-sast-enhancements) [Secrets Detection & Prevention](https://www.legitsecurity.com/enterprise-secret-scanning) [](https://www.legitsecurity.com/enterprise-secret-scanning) [Software Supply Chain Security](https://www.legitsecurity.com/software-supply-chain-security) [](https://www.legitsecurity.com/software-supply-chain-security) [Advanced Code Change Management](https://www.legitsecurity.com/advanced-code-change-management-in-software-development-for-safe-releases) [](https://www.legitsecurity.com/advanced-code-change-management-in-software-development-for-safe-releases) [Compliance](https://www.legitsecurity.com/continuous-compliance-sbom) [](https://www.legitsecurity.com/continuous-compliance-sbom)
* Secure AI Code Secure AI Code & Apps
Overview
AI is revolutionizing development - making it faster, smarter, and more autonomous. It’s also rewriting the rules of application security. Traditional AppSec tools weren’t designed for AI-driven dev processes. Legit is here to help.
[Learn More](https://www.legitsecurity.com/ai-native-application-security-platform-for-modern-sdlc) Use Case [VibeGuard New](https://www.legitsecurity.com/security-governance-for-ai-generated-code-legit-vibeguard) [](https://www.legitsecurity.com/security-governance-for-ai-generated-code-legit-vibeguard) [MCP Server](https://www.legitsecurity.com/legit-mcp-server-ai-native-security-intelligence-for-developers) [](https://www.legitsecurity.com/legit-mcp-server-ai-native-security-intelligence-for-developers) [AI-Powered Remediation](https://www.legitsecurity.com/ai-powered-remediation-fix-appsec-issues-faster-with-legit) [](https://www.legitsecurity.com/ai-powered-remediation-fix-appsec-issues-faster-with-legit) [AI Visibility](https://www.legitsecurity.com/ai-visibility) [](https://www.legitsecurity.com/ai-visibility)
* Resources [Blog](https://www.legitsecurity.com/blog) [](https://www.legitsecurity.com/blog) [Resource Library](https://www.legitsecurity.com/resource-library) [](https://www.le
*[Content truncated...]*
---
## 7. New framework simplifies the complex landscape of agentic AI
**URL:** [https://venturebeat.com/orchestration/new-framework-simplifies-the-complex-landscape-of-agentic-ai](https://venturebeat.com/orchestration/new-framework-simplifies-the-complex-landscape-of-agentic-ai)
**Score:** 0.9964
### Summary
As the AI landscape matures, the focus is shifting from building one giant, perfect model to constructing a smart ecosystem of specialized tools around a stable core. For most enterprises, the most effective path to agentic AI isn't building a bigger brain but giving the brain better tools.
##### Subscribe to get latest news!
Deep insights for enterprise AI, data, and security leaders
VB Daily
AI Weekly
AGI Weekly
Security Weekly
Data Infrastructure Weekly
VB Events
All of them [...] By submitting your email, you agree to ourTermsandPrivacy Notice.
Get updates
You're in! Our latest news will be hitting your inbox soon.
More
[](
[](
Press Releases
Contact Us
Advertise
Share a News Tip
Contribute
Privacy Policy
Terms of Service
Do not sell my personal info
© 2025 VentureBeat. All rights reserved. [...] A prime example is DeepSeek-R1, where the model was trained through reinforcement learning with verifiable rewards to generate code that successfully executes in a sandbox. The feedback signal is binary and objective (did the code run, or did it crash?). This method builds strong low-level competence in stable, verifiable domains like coding or SQL.
### Full Content
New framework simplifies the complex landscape of agentic AI | VentureBeat
===============
[](https://venturebeat.com/)
* [Orchestration](https://venturebeat.com/category/orchestration)
* [Infrastructure](https://venturebeat.com/category/infrastructure)
* [Data](https://venturebeat.com/category/data)
* [Security](https://venturebeat.com/category/security)
More
[Newsletters](https://venturebeat.com/newsletters)
[All Posts](https://venturebeat.com/)
New framework simplifies the complex landscape of agentic AI
============================================================
[Ben Dickson](https://venturebeat.com/author/ben-dickson-techtalks) December 29, 2025
Image credit: VentureBeat with ChatGPT
[](https://www.google.com/preferences/source?q=venturebeat.com "Add to Google Preferred Source")
With the ecosystem of agentic tools and frameworks exploding in size, navigating the many options for building AI systems is becoming increasingly difficult, leaving developers confused and paralyzed when choosing the right tools and models for their applications.
In a [new study](https://arxiv.org/abs/2512.16301), researchers from multiple institutions present a comprehensive framework to untangle this complex web. They categorize agentic frameworks based on their area of focus and tradeoffs, providing a practical guide for developers to choose the right tools and strategies for their applications.
For enterprise teams, this reframes agentic AI from a model-selection problem into an architectural decision about where to spend training budget, how much modularity to preserve, and what tradeoffs they’re willing to make between cost, flexibility, and risk.
Agent vs. tool adaptation
-------------------------
The researchers divide the landscape into two primary dimensions: **agent adaptation** and **tool adaptation**.
Agent adaptation involves modifying the foundation model that underlies the agentic system. This is done by updating the agent’s internal parameters or policies through methods like fine-tuning or reinforcement learning to better align with specific tasks.
Tool adaptation, on the other hand, shifts the focus to the environment surrounding the agent. Instead of retraining the large, expensive foundation model, developers optimize the external tools such as search retrievers, memory modules, or sub-agents. In this strategy, the main agent remains "frozen" (unchanged). This approach allows the system to evolve without the massive computational cost of retraining the core model.
Agentic adaptation strategies (source: arXiv)
The study further breaks these down into four distinct strategies:
**A1: Tool execution signaled:** In this strategy, the agent learns by doing. It is optimized using verifiable feedback directly from a tool's execution, such as a code compiler interacting with a script or a database returning search results. This teaches the agent the "mechanics" of using a tool correctly.
A prime example is [DeepSeek-R1](https://venturebeat.com/ai/deepseek-r1-is-a-boon-for-enterprises-making-ai-apps-cheaper-easier-to-build-and-more-innovative), where the model was trained through reinforcement learning with verifiable rewards to generate code that successfully executes in a sandbox. The feedback signal is binary and objective (did the code run, or did it crash?). This method builds strong low-level competence in stable, verifiable domains like coding or SQL.
**A2: Agent output Signaled:**Here, the agent is optimized based on the quality of its final answer, regardless of the intermediate steps and number of tool calls it makes. This teaches the agent how to orchestrate various tools to reach a correct conclusion.
An example is [Search-R1](https://venturebeat.com/ai/beyond-rag-search-r1-integrates-search-engines-directly-into-reasoning-models), an agent that performs multi-step retrieval to answer questions. The model receives a reward only if the final answer is correct, implicitly forcing it to learn better search and reasoning strategies to maximize that reward. A2 is ideal for system-level orchestration, enabling agents to handle complex workflows.
**T1: Agent-agnostic:** In this category, tools are trained independently on broad data and then "plugged in" to a frozen agent. Think of classic dense retrievers used in RAG systems. A standard retriever model is trained on generic search data. A powerful frozen LLM can use this retriever to find information, even though
*[Content truncated...]*
---
## 8. The agentic reality check: Preparing for a silicon-based ...
**URL:** [https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends/2026/agentic-ai-strategy.html](https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends/2026/agentic-ai-strategy.html)
**Score:** 0.9960
### Summary
9. Marie Myers (executive vice president and chief financial officer, HPE), interview with Deloitte, March 1, 2025.
10. John Roese (chief technology officer and chief AI officer, Dell Technologies), interview with Deloitte, Sept. 29, 2025.
11. Maribel Solanas Gonzalez (group chief data officer, Mapfre Insurance), interview with Deloitte, June 18, 2024.
12. "Reimagining operations with agentic AI at Toyota,” _\_Deloitte Insights\__, Dec. 3, 2025. [...] 17. AgentCommunicationProtocol.dev, “Welcome,” accessed Nov. 6, 2025.
18. Saad Merchant, “ACP: Future of offline AI agent collaboration,” Alumio, Oct. 24, 2025.
19. Kearney, “FinOps for AI and AI for FinOps,” Jan. 28, 2025.
20. Jake Latimer, “Will AI be taxed? The debate over AI-powered businesses: The 2025 tech-tax tussle,”_Medium_, March 13, 2025.
21. Ken Huang, “Agentic AI identity management approach,” Cloud Security Alliance, March 11, 2025. [...] 13. Tracey Franklin (chief people and digital technology officer, Moderna), interview with Deloitte, Sept. 26, 2025.
14. Aditya Challapally, Chris Pease, Ramesh Raskar, and Pradyumna Chari, “The gen AI divide: State of AI in business 2025,” July 2025.
15. Anthropic, PBC, “Introducing the model context protocol,” Nov. 25, 2024.
16. Rao Surapaneni, Miku Jha, Michael Vakoc, and Todd Segal, “Announcing the Agent2Agent Protocol (A2A),” Google for Developers, April 9, 2025.
### Full Content
Agentic AI strategy | Deloitte Insights
===============
Deloitte Insights and our research centers deliver proprietary research designed to help organizations turn their aspirations into action.
#### DELOITTE INSIGHTS
* [Home](https://www.deloitte.com/us/en/insights.html?icid=disidenav_insights)
* [Spotlight](https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends/2026/agentic-ai-strategy.html#)
* [Weekly Global Economic Outlook](https://www.deloitte.com/us/en/insights/topics/economy/global-economic-outlook/weekly-update.html?icid=disidenav_weekly-update)
* [Tech Trends](https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends.html?icid=disidenav_tech-trends)
* [Human Capital Trends](https://www.deloitte.com/us/en/insights/topics/talent/human-capital-trends.html?icid=disidenav_human-capital-trends)
* [Digital Media Trends](https://www.deloitte.com/us/en/insights/industry/technology/digital-media-trends-consumption-habits-survey.html?icid=disidenav_digital-media-trends-consumption-habits-survey)
* [TMT Predictions](https://www.deloitte.com/us/en/insights/industry/technology/technology-media-and-telecom-predictions.html?icid=disidenav_technology-media-and-telecom-predictions)
* [FSI Predictions](https://www.deloitte.com/us/en/insights/industry/financial-services/financial-services-industry-predictions.html?icid=disidenav_financial-services-industry-predictions)
* [Topics](https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends/2026/agentic-ai-strategy.html#)
* [Economics](https://www.deloitte.com/us/en/insights/topics/economy.html?icid=disidenav_economy)
* [Environmental, Social, & Governance](https://www.deloitte.com/us/en/insights/topics/environmental-social-governance.html?icid=disidenav_environmental-social-governance)
* [Operations](https://www.deloitte.com/us/en/insights/topics/operations.html?icid=disidenav_operations)
* [Strategy](https://www.deloitte.com/us/en/insights/topics/business-strategy-growth.html?icid=disidenav_business-strategy-growth)
* [Technology](https://www.deloitte.com/us/en/insights/topics/technology-management.html?icid=disidenav_technology-management)
* [Workforce](https://www.deloitte.com/us/en/insights/topics/talent.html?icid=disidenav_talent)
* [Industries](https://www.deloitte.com/us/en/insights/industry.html?icid=disidenav_industry)
* [More](https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends/2026/agentic-ai-strategy.html#)
* [About](https://www.deloitte.com/us/en/insights/about.html?icid=disidenav_about)
* [Deloitte Insights Magazine](https://www.deloitte.com/us/en/insights/deloitte-insights-magazine.html?icid=disidenav_deloitte-insights-magazine)
* [Top 10 Reading Guide](https://www.deloitte.com/us/en/insights/top-10-business-insights.html?icid=disidenav_top-10-business-insights)
* [Videos](https://www.deloitte.com/us/en/insights/multimedia/videos.html?icid=disidenav_videos)
#### DELOITTE RESEARCH CENTERS
* [Cross-Industry](https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends/2026/agentic-ai-strategy.html#)
* [Home](https://www.deloitte.com/us/en/insights/research-centers/center-for-integrated-research.html?icid=disidenav_center-for-integrated-research)
* [Workforce Trends](https://www.deloitte.com/us/en/insights/research-centers/center-for-integrated-research/workforce-trends.html?icid=disidenav_workforce-trends)
* [Enterprise Growth & Innovation](https://www.deloitte.com/us/en/insights/research-centers/center-for-integrated-research/enterprise-growth-and-innovation.html?icid=disidenav_enterprise-growth-and-innovation)
* [Technology & Transformation](https://www.deloitte.com/us/en/insights/research-centers/center-for-integrated-research/technology-and-transformation.html?icid=disidenav_technology-and-transformation)
* [Environmental & Social Issues](https://www.deloitte.com/us/en/insights/research-centers/center-for-integrated-research/environmental-and-social-issues.html?icid=disidenav_environmental-and-social-issues)
* [Economics](https://www.deloitte.com/us/en/insights/topics/technology-management/tech-trends/2026/agentic-ai-strategy.html#)
* [Home](https://www.deloitte.com/us/en/insights/research-centers/economics.html?icid=disidenav_economics)
* [Consumer Spending](https://www.deloitte.com/us/en/insights/research-centers/economics/consumer-spending.html?icid=disidenav_consumer-spending)
* [Housing](https://www.deloitte.com/us/en/insights/research-centers/economics/housing.html?icid=disidenav_housing)
* [Business Investment](https://www.deloitte.com/us/en/insights/research-centers/economics/business-investment.html?icid=disidenav_business-investment)
* [Globalization & International Trade](https://www.deloitte.com/us/en/insights/research-centers/economics/gl
*[Content truncated...]*
---