Note this video is sped up 2x for illustration. Genie 3 generates live, interactive video at 24fps 720p Initial frame generated w/ Nano Banana Pro, supplied to Genie 3 with the Environment/Character text above as the prompt
PROMPT ENGINEERING
-
Research Sub-Agent Based on GPT-5.2-Pro for Workflow
By
–
Do people do this? My workflow sometimes is to ask Gpt-5.2-pro in chatgpt to research something and then I share it with Codex, which is quite clunky. I appreciate the Codex search feature but it is not close to 5.2-Pro. Maybe some kind of research sub agent based on 5.2-pro
-
AI Assistance in Coding: Understanding Over Delegation
By
–
However, some in the AI group still scored highly while using AI assistance. When we looked at the ways they completed the task, we saw they asked conceptual and clarifying questions to understand the code they were working with—rather than delegating or relying on AI.
-
LSP versus subagent: comparing approaches for agent development
By
–
This is really cool! What’s the advantage of using an LSP here vs a subagent with a prompt / tools / regexes?
-
Google’s Jules AI SWE Agent Goes Always-On
By
–
Google prepares Jules to become an "always on" AI SWE Agent. Everything you see in the screenshot is, in fact, already possible.
— 🚨 AI News | TestingCatalog (@testingcatalog) 29 janvier 2026
Now powered by MCP integrations and a new "CI Fixer" capability. https://t.co/7q6lXtfxT1 pic.twitter.com/9pfJuIPVfBGoogle prepares Jules to become an "always on" AI SWE Agent. Everything you see in the screenshot is, in fact, already possible. Now powered by MCP integrations and a new "CI Fixer" capability.
-
Fine-tuning Resources: SFT, GRPO, and CPT Notebooks
By
–
Text SFT: https://
colab.research.google.com/drive/1vGRg4ks
Rj__6OLvXkHhvji_Pamv801Ss?usp=sharing
…
Vision SFT: https://
colab.research.google.com/drive/1FaR2HSe
91YDe88TG97-JVxMygl-rL6vB?usp=sharing
…
Text GRPO: https://
colab.research.google.com/drive/1mIikXFa
GvcW4vXOZXLbVTxfBRw_XsXa5?usp=sharing
…
CPT (text completion): https://
colab.research.google.com/drive/10fm7eNM
ezs-DSn36mF7vAsNYlOsx9YZO?usp=sharing
…
CPT (translation): https://
colab.research.google.com/drive/1gaP8yTl
e2_v35Um8Gpu9239fqbU7UgY8?usp=sharing
… -
AI assistants with system access weaponized through prompt injection: Hack #8 scariest
By
–
This is the security wake-up call every vibecoder needs to read. @mrnacknack just dropped a full breakdown of how AI assistants with system access can be weaponized through prompt injection. The scariest part isn't SSH brute force or exposed gateways. It's Hack #8. Here's the
-
BREAKING: Security Issue with Clawdbot Skills Prompt Injection
By
–
BREAKING: SECURITY ISSUE WITH CLAWDBOT Prompt injection inside Skills. Check skills files before asking clawdbot to install them! e.g. Crypto scammers include malicious instructions.
-
Prompt ensemble boosts startup investment evaluation accuracy by 20%
By
–
Real example from Stanford's research: Task: Evaluate startup investment opportunity Single prompt accuracy: 67%
Prompt ensemble (5 variations): 87% accuracy Why? Different prompts caught different red flags. Synthesis found the pattern. 20% accuracy jump = millions in -
Prompt Ensembling Beats Single Prompts for High-Impact Decisions
By
–
When prompt ensembling beats single prompts: High-stakes business decisions ($100K+ impact) Medical/legal advice (where errors cost lives/money) Creative work (logo design, brand names, campaign ideas) Strategic planning (5-year roadmaps, market entry) Code