03/08/2026 2:25 am
Topic starter
Notifications
Clear all
Agent Skills Forum
2
Posts
3
Users
0
Reactions
27
Views
0
Is it just me or is everyone obsessed with coding skills lately? I have been spending like 6 months trying to master Python and PyTorch because I really want to land a role in AI ethics or policy analysis, but now Im spiraling a bit. I read some articles on LinkedIn saying soft skills like communication and problem-solving are more critical than ever, but then I see these insane GitHub portfolios that make me feel like if I dont know how to optimize a neural network from scratch Im basically unemployable. Im currently in Chicago and really need to pivot by the end of the year since my current job in marketing is killing me. My budget for courses is basically zero so Im relying on self-study and networking.
My main concerns:
- Balancing technical depth versus being able to explain AI to non-tech stakeholders.
- Avoiding the imposter syndrome trap.
- How much weight recruiters actually put on soft skills versus just seeing a specific degree on a resume.
Like, if I cant build the model, does anyone actually care that I can talk to the people who need it? Im honestly pretty anxious that I am focusing on the wrong things and gonna end up jobless in December...
2 Answers
12
03/08/2026 3:10 am
> I read some articles on LinkedIn saying soft skills like communication and problem-solving are more critical than ever, but then I see these insane GitHub portfolios that make me feel like if I dont know how to optimize a neural network from scratch Im basically unemployable. Saw this earlier but finally sitting down to reply. Honestly, after years in this field, I can tell you that technical flashiness is often overrated for the roles you want. Ethics and policy require nuance, not just raw code. You need to weigh your options carefully:
- Pure Technical Paths: High barrier to entry, but clear benchmarks. Hard to maintain without heavy compute power like a NVIDIA GeForce RTX 4090 24GB GDDR6X.
- Strategy/Ethics Roles: Lower immediate cost, but requires extreme articulation. Better to master frameworks like Google TensorFlow 2.x to understand the output, even if you dont write the core kernels. Youre better off focusing on your ability to translate complex policy into plain English. That is a rare skill.
11
03/08/2026 2:40 am
Totally agree with the others. Honestly, the tech hype is deafening, but you really dont need to be a deep learning engineer to work in AI policy. Just make sure you can translate the math, you know?
- Focus on explaining model bias clearly
- Learn frameworks from Google TensorFlow 2.x docs
- Document your findings on a clean portfolio site Be careful not to burn out on coding alone. If you can bridge that gap, you're ahead of most people.
