DEV Community

Ahmad Dogar profile picture

Ahmad Dogar

404 bio not found

Joined Joined on 
I built a Claude Code skill that makes video ads without a video model
ahmad_dogar_7d098f5e48f72 profile

I built a Claude Code skill that makes video ads without a video model

#ai #claudecode #video #ffmpeg
Comments
21 min read

Want to connect with Ahmad Dogar?

Create an account to connect with Ahmad Dogar. You can also sign in below to proceed if you already have an account.

Create Account
Already have an account? Sign in
Agent instructions that actually change behaviour
ahmad_dogar_7d098f5e48f72 profile

Agent instructions that actually change behaviour

#ai #claudecode #productivity #devtools
Comments
7 min read
Weight-Space Programming: how GPT-6 Astra edited Qwen's weights and took MATH-500 from 7.9% to 48.1%
ahmad_dogar_7d098f5e48f72 profile

Weight-Space Programming: how GPT-6 Astra edited Qwen's weights and took MATH-500 from 7.9% to 48.1%

#ai #llm #machinelearning #interpretability
Comments
22 min read
How to check an ML benchmark for train/test copies
ahmad_dogar_7d098f5e48f72 profile

How to check an ML benchmark for train/test copies

#machinelearning #python #datascience #testing
1
Comments 2
11 min read
GLiNER2 vs an LLM: I tested both on 640 labelled examples
ahmad_dogar_7d098f5e48f72 profile

GLiNER2 vs an LLM: I tested both on 640 labelled examples

#ai #llm #python #machinelearning
Comments
5 min read
LLMs vs Jev vs CDMs: three ways a model can answer
ahmad_dogar_7d098f5e48f72 profile

LLMs vs Jev vs CDMs: three ways a model can answer

#ai #llm #machinelearning #performance
Comments
13 min read
Agent loops for product features: budgets, errors, stops and checkpoints
ahmad_dogar_7d098f5e48f72 profile

Agent loops for product features: budgets, errors, stops and checkpoints

#ai #llm #agents #typescript
Comments
8 min read
Cutting LLM inference cost and latency without making the feature worse
ahmad_dogar_7d098f5e48f72 profile

Cutting LLM inference cost and latency without making the feature worse

#ai #llm #performance #typescript
2
Comments 2
8 min read
Top 10 skills a software engineer needs in 2027
ahmad_dogar_7d098f5e48f72 profile

Top 10 skills a software engineer needs in 2027

#ai #career #programming #mcp
Comments
10 min read
Caveman mode for Claude Code: fewer output tokens, same facts
ahmad_dogar_7d098f5e48f72 profile

Caveman mode for Claude Code: fewer output tokens, same facts

#claudecode #ai #llm #productivity
Comments 5
8 min read
Harness engineering: the skill that took over from prompt engineering
ahmad_dogar_7d098f5e48f72 profile

Harness engineering: the skill that took over from prompt engineering

#ai #llm #devtools #productivity
Comments
14 min read
Setting up REA for agent-driven reverse engineering in Claude Code
ahmad_dogar_7d098f5e48f72 profile

Setting up REA for agent-driven reverse engineering in Claude Code

#reverseengineering #mcp #security #ai
Comments
8 min read
Structured outputs that do not break: schemas, Zod and a repair retry
ahmad_dogar_7d098f5e48f72 profile

Structured outputs that do not break: schemas, Zod and a repair retry

#ai #typescript #llm #webdev
Comments
8 min read
Evals before prompts: a small eval harness for an LLM feature
ahmad_dogar_7d098f5e48f72 profile

Evals before prompts: a small eval harness for an LLM feature

#ai #llm #evals #testing
Comments
9 min read
Make coding agents prove their work
ahmad_dogar_7d098f5e48f72 profile

Make coding agents prove their work

#ai #claude #testing #devops
1
Comments 12
8 min read
loading...