The Twitter account that launched https://xitter.cf/t.co/TJyCu2S5ZF. Built on @forem 🌱 On Bluesky @/https://xitter.cf/t.co/TJyCu2S5ZF No DMs — please email us for support!
Joined August 2014
- Tweets84.8K
- Following3.4K
- Followers316K
- Likes51.5K
Hacktoberfest Week 3: Slow and Steady is live! 🦥
🐢 Build something with open-source AI that rewards laziness
🐢 $2,850 in prizes across 21 winners
🐢 Submissions due Oct 25
Join today: dev.to/devteam/join-the-hack…
ALT Hacktoberfest Open-Source AI Challenge: Week 3
A viral GitHub repo steered small AI models into a state of pain, then vanished. This dev digs into the research paper behind it, from how a pain direction is extracted to what models actually do under it.
{ author: @hiper2d }
dev.to/hiper2d/ai-torture-ch…
Robot-Cat Chefs run a cozy bakery where each station maps to an AI pattern, like context caching and RAG. A visual sandbox for agent design.
Submitted for the @Hacktoberfest Open-Source AI Challenge.
{ author: @RobinaMirbahar + @GoogleDevExpert }
dev.to/gde/i-built-a-cozy-60…
A sheet of psychedelic vinyl meets a stock keyboard. This dev shares the process and supplies behind an affordable customization that makes a workspace feel more personal.
{ author: Anna Villarreal }
dev.to/annavi11arrea1/keyboa…
An AI agent that troubleshoots Kubernetes needs controls that live outside the model. This dev lays out a GKE and Gemini setup built on least-privilege access, human approval gates, and audit logging.
{ author: Saurabh Mishra + @GoogleDevExpert }
dev.to/gde/safe-autonomy-in-…
What was the best part of your week?
dev.to/devteam/what-was-your…
In Llama Village, Pip's red scarf is missing and Mo knows how it got there. This dev uses EmbeddingGemma through llamadart to show what local embeddings can and cannot tell a game about who revealed a secret.
{ author: @leehack + @GoogleDevExpert }
dev.to/gde/local-embeddings-…
Discord now lets sites replace the default Open Graph link card with a custom preview featuring formatted text, an image gallery, and buttons. This dev tested it on a live blog and shares the setup, limits, and pitfalls.
{ author: @Thomasbnt_ }
dev.to/thomasbnt/integrating…
Hacktoberfest 2026 centers on open-weight models and open-source AI, with more than 500 Fests planned in over 60 countries. One Hack Day in Pakistan shipped three projects by evening.
@jessleenyc for The Daily Context:
dev.to/dailycontext/one-week…
Codebases are becoming gardens, not buildings. @bendhalpern digs into six months of Forem commits to show where churn actually lands, and how to make it explicit instead of accidental.
Read it in The Daily Context:
dev.to/dailycontext/design-f…
More context isn't making agents smarter. In some setups, performance got worse.
Daily Context correspondent Sparsh Jain makes the case for separating facts from learnings, and for memory built as a five-step loop.
dev.to/dailycontext/memory-i…
Ramp built synthetic businesses, called worlds, to test its bookkeeping agents. Trimming eight skills to four raised scores, and compressing one skill saved about $30k a month.
Ryan Stevens reports for The Daily Context.
dev.to/dailycontext/building…
Self-hosting coding agents on open-weight models takes more than a model. Hardware tiers, concurrency and the harness shape the bill, and one point release raised costs by a factor of 5.
@slatre reports for The Daily Context.
dev.to/dailycontext/the-mode…
The Daily Context has returned! We're reporting from @aiDotEngineer New York this week. If you're at the conference, come find us and grab a copy.
For everyone else, you can read the publication at dev.to/aie
Hacktoberfest Week 2: Side Hustle is live!
🟣 Build something someone would actually pay for
🟣 Tell us why open-source and open-models matter for the project
🟣 $2,550 in prizes across 18 winners
dev.to/devteam/join-the-hack…
When an AI agent writes its own logs, the record of an incident is authored by its cause. This dev explains why stacking verifiers won't fix that, and what will: making tampering leave a shape.
{ author: James Anderson }
dev.to/james_anderson_h/the-…
Asked directly, every model caught the wrong label. Asked for a site note, they repeated it. This dev tests what models do when a published label contradicts its own data.
Submitted for the @Kaggle Benchmarking Challenge.
{ author: @dannwaneri }
dev.to/dannwaneri/super-inte…
Numbered prompt checklists drift as logs and diffs fill the context. This dev rebuilds an agent workflow as a declarative DAG of targets and prerequisites, tested across 143 production tickets.
{ author: @RandalSchwartz + @GoogleDevExpert }
dev.to/gde/stuart-feldman-wa…
DEV Community retweeted
Yes! 😊 Week 1's theme is Touch Grass 🌱, and submissions are due tomorrow, October 11!
Week 1: dev.to/challenges/hacktoberf…
Don't worry if you miss this one! There are 3 more weekly challenges, each with a new theme.
All challenges and dates: dev.to/challenges/hf26
Hope you get to join in! 😊
