Bilingual

You're Not a Better Engineer Because You Type Git Commands by Hand
11 days ago
- AI agents handle tedious engineering tasks like commit messages, PR descriptions, and code comments, saving significant time.
- AI ensures consistency in naming conventions, Git hygiene, issue tracking, and release workflows, reducing errors and conflicts.
- LLMs benefit from structured codebase context (comments, rules, paths), enabling more fluid AI-assisted interactions and challenging inconsistencies.
- Agents manage Git operations (branch creation, PR preparation, label updates) and automate delivery checks, turning rituals into efficient sequences.
- Delegating repetitive tasks to AI prevents burning money and life on mechanical steps, allowing focus on solving actual problems.
Can AI build a jet engine? JARVIS Challenge tests AI copilots
5 days ago
- The JARVIS Challenge at MIT tested if AI can accelerate the design-build-test cycle for a small gas turbine engine, with undergraduate teams building engines in four weeks.
- AI tools, accessed via the Parley platform, helped with tasks like summarizing textbooks, software tutorials, vendor sourcing, and design trade studies, but faced limitations like hallucinations and lack of physical understanding.
- Teams found that engineering judgment and expertise were crucial; while AI assisted, human oversight was needed to trust, challenge, and translate AI outputs into working hardware.
- Younger students used AI more frequently, while senior teams relied on deeper experience; the winning team was initially resistant to AI, trusting fundamentals and teamwork instead.
- Manufacturing and vendor relationships emerged as rate-limiting steps, with personal connections proving more reliable than AI-sourced vendors for tight timelines.
- The challenge demonstrated that AI can compress engineering cycles, but education and hands-on experience remain vital for developing the judgment to effectively lead AI tools.
AI, Vim, and the Illusion of Flow
19 hours ago
- Vim enhances flow by making text editing seamless, not by increasing speed, and it directly translates user intent into output without distortion.
- AI tools promise similar fluidity but introduce a 'depth problem': their output appears polished but may not accurately reflect the operator's intent or skill, making it hard to spot flaws in code.
- The gap between 'looking done' and 'being right' is growing, as AI prototypes lack edge cases, error handling, and other critical details, leading to unrealistic expectations.
- AI fosters an 'empathy gap' where surface-level contributions across disciplines create false confidence in understanding others' jobs, undermining organizational respect for depth.
- To address these issues, we need to set clear boundaries for AI output, invest in education on evaluating AI work, build cross-disciplinary empathy, and protect personal flow by ensuring AI enhances rather than replaces problem-solving.

Hasty Briefsbeta

#ai in engineering

You're Not a Better Engineer Because You Type Git Commands by Hand

Can AI build a jet engine? JARVIS Challenge tests AI copilots

AI, Vim, and the Illusion of Flow