The missing layer in AI-assisted development We have SBOMs for dependencies and signed commits for authorship. Why don't we have provenance for AI contributions? Here's a thing that's been bothering me. Every engineering team I talk to is usin...

We write a lot of #AI code, but reviewing it has become a bottleneck.
I came up with AI Code Provenance (ACP): a draft standard & ref implementation for tracking intent behind AI-generated code.

markan.me/the-missing-...

#codegen #claudecode #cursor #codex

Amplifying — AI Benchmark Research Systematic analysis of how AI systems make decisions — from product recommendations to developer tool choices.

What Claude Code Actually Chooses:
In this study, they pointed Claude Code at real repos 2,430 times and watched what it chose. No tool names in any prompt. Open-ended questions only. Insightful read: amplifying.ai/research/cla... #claude #ai #codegen

Как мы написали React Native библиотеку для Яндекс Карт за два дня с Claude Сначала коротко о том, зачем нам это было ...

#React #Native #Яндекс #Карты #Fabric #TurboModules #Codegen #Claude #iOS #Android #нативные

Origin | Interest | Match

📊 Por qué ya no evaluamos con SWE-bench Verified

Contaminación y medición errónea del progreso en código frontera.

openai.com/index/why-we-no-longer-e...

#Benchmarking #AIEngineering #CodeGen #RoxsRoss

Release Goa v3.25.0 · goadesign/goa Highlights Automatic HTTP route tagging: the default Goa muxer now sets r.Pattern on every dispatched request, enabling observability middleware like otelhttp (v0.65.0+) to tag spans and metrics w...

Goa v3.25.0 is out!

→ Automatic route tagging via r.Pattern — otelhttp picks up routes with zero config
→ DSL errors now show file:line locations
→ Mixed unary + SSE on one method
→ Bug fixes

github.com/goadesign/goa/releases/tag/v3.25.0

#golang #opentelemetry #opensource #codegen

Vibe coding reminds me of (Macromedia) Dreamweaver. The result worked, but the generated code was trash. It’s ok to use genAI as an accelerator, but make sure that you understand every single line #genai #codegen #llm #novibecoding

However, others note LLVM IR is increasingly diverging from C/C++ semantics. This forces continuous adaptation for developers, increasing the effort to maintain projects that generate custom IR. #CodeGen 3/7

The Emperor Has No Clothes: How to Code Claude Code in 200 Lines of Code The core of tools like Claude Code, Cursor, and Warp isn't magic. It's about 200 lines of straightforward Python. Let's build one from scratch.

If you use Claude Code or Cursor or other AI code-gen tools, this is a good way to understand code generation from first principles (also, gives a practical illustration of tool calling and the basics of an agent): www.mihaileric.com/The-Emperor-... #ai #llms #codegen #developers

Funding Raises: #7AI $130M Series A; #ChaiDiscovery $130M Series B; #ProtegoBiopharma $130M Series B; #Mythic $125M; #AtavistikBio & #TrianaBiomedicines each $120M Series B; #Harmonic $120M Series C

Acquisitions: #Anysphere > #Graphite; #ClickUP > #Codegen; #Monzo > #Habito

Highlights from over a year of Cursor usage… #cursor #ai #codegen

The Population Test Your Agent Must Pass | Maria Gorinova YouTube video by AI Native Dev

Learn more about our work on abstraction adherence of coding agents! www.youtube.com/watch?v=nx5_...

#coding #aicoding #agenticai #codegen #devtools #cursor #claudecode

Interesting claim from Nadella: 20-30% of Microsoft's code now AI-generated. Skeptics question the metrics amid fuzzy human-AI authorship lines. What does it mean for dev productivity and software engineering? #AI #CodeGen

Compile SQL to type-safe code Compile SQL to type-safe code

Just got finished setting up sqlc tooling for a monorepo with Go and Python. Now it will generate code for each language based on a single set of *.sql files with query statements.

- sqlc.dev
- github.com/sqlc-dev/sql...

#GoLang #Python #PythonLang #sqlc #CodeGen

#CodeGen #AIcoding #DeveloperTools #AIAssistant #CodeAutomation #Programming #SoftwareDevelopment #MachineLearning

Identifying Accessibility Data Gaps in CodeGen Models :: Aaron Gustafson Late last year, I probed an LLM’s responses to HTML code generation prompts to assess its adherence to accessibility best practices. The results were unsurprisingly disappointing — roughly what I’d…

Aaron Gustafson highlights accessibility gaps in AI-generated code, showing the need for better training to ensure inclusive digital experiences.

www.aaron-gustafson.com/notebook/ide...

#AIandAccessibility #WebDevelopment #CodeGen #MachineLearning #DigitalAccessibility #AIethics #A11y

🎉 Goa v3.22.6 is out!

New: Better SSE encoding with primitive-friendly formatting
Fixed: HTTP/2 retry errors, interceptor attribute validation, OpenAPI wildcard paths, and CLI escaping

Release notes: github.com/goadesign/goa/releases/tag/v3.22.6

#golang #api #codegen #golang #microservice

Identifying Accessibility Data Gaps in CodeGen Models :: Aaron Gustafson Late last year, I probed an LLM’s responses to HTML code generation prompts to assess its adherence to accessibility best practices. The results were unsurprisingly disappointing — roughly what I’d…

AI tools still struggle with accessible code.
Aaron Gustafson digs into where CodeGen models miss the mark and what we can do about it.

www.aaron-gustafson.com/notebook/ide...

#Accessibility #AI #CodeGen #InclusiveTech #A11y #WebDev #DigitalAccessibility

Comparative Study of AI Code Generation Assistants' Security

A study of 61 code snippets from GPT‑4, GPT‑3.5, Google Bard and Gemini found GPT‑4 had the highest functional correctness, and security‑aware prompts reduced vulnerabilities. Read more: getnews.me/comparative-study-of-ai-... #ai #codegen

GPT‑OSS 20B Outperforms 120B Model in Code and Benchmark Tests

In August 2025 OpenAI released the GPT‑OSS family; the 20 billion‑parameter model outperformed the 120 billion‑parameter version on HumanEval and MMLU benchmarks, using less memory and energy. getnews.me/gpt-oss-20b-outperforms-... #gptoss #codegen

Let Playwright Codegen capture your clicks and MailSlurp validate the inbox—newsletter tests without a single line of code.

#PlaywrightJS #Codegen #NoCodeTesting #EmailQA #BrowserAutomation #MailSlurp

www.youtube.com/watch?v=cvQF...

Anthropic's Claude 3.7 Sonnet is the new king 👑 of code generation (but only with help), and DeepSeek R1 disappoints buff.ly/BUvGPLL
#Java #CodeGen #genai #llm

Webflow jumps into AI coding 🚀
From prompt to production in one platform.
With AEO, it’s not just SEO anymore—it’s Answer Engine Optimization.

#WebflowAI #AEO #AI #CodeGen #SEO

Evaluating LLMs for Functional Code in Industrial Environments

A case study at a semiconductor equipment maker introduced the build@k metric, which counts compilation in first k attempts, and found few-shot and chain-of-thought prompts boost success. getnews.me/evaluating-llms-for-func... #llm #codegen #semiconductor

FGIT Technique Boosts Code Generation Accuracy in LLMs

Fault-Guided Fine-Tuning (FGIT) boosts code generation, delivering a 6.9% average pass@1 improvement and letting 6.7B-parameter models outperform GPT-3.5-Turbo on benchmarks. Read more: getnews.me/fgit-technique-boosts-co... #fgit #codegen #llm

I’ve made a career building code gen and low / no code tools. I get how AI based code generators and the other tools work. I thought the field was crowded. But if you niche-down, I think there’s still plenty of opportunities.

#scriptable #locode #lowcode #nocode #codegen #cursor

Release v3.22.2 · goadesign/goa Goa v3.22.2 Release Notes Highlights Prevent invalid code when different declared types share the same TypeName. Goa now validates duplicate TypeNames across declared ResultTypes and UserTypes (ge...

Just shipped Goa v3.22.2 🎉

- Safer types (duplicate TypeName checks)
- JSON‑RPC: ID + streaming fixes
- HTTP/SSE: Last-Event-ID fix
- gRPC: better OneOf/Any
No breaking changes.

Notes: github.com/goadesign/go...
#golang #goa #microservices #openapi #codegen #jsonrpc #grpc

Kate Marshalkina - At the top of the pyramid: Playwright testing at scale In the past year, we went from zero to 20,000 lines of Playwright tests, pushing its capabilities beyond basic end-to-end testing. This talk will explore the challenges we encountered as we scaled our Playwright setup and how we solved them, thanks to the robust ecosystem and team creativity.

From 0 to 20k #Playwright #testing lines in a year 🚀 This talk dives into the challenges of scaling beyond basic #E2E, and how ecosystem strength + team creativity got us through

Watch this amazing talk by @kalabro.tech
youtu.be/JZPe4fsKxS8

#ArtificialIntelligence #AI #Codegen