Lim Chee Kin limcheekin

Spot the Flaws Prompt: Act as a recruiter for [your industry/role]. Review my resume below and highlight weak areas, overused buzzwords, and missing metrics. Be brutally honest.
Rewrite for Impact Prompt: Rewrite this resume to sound more results-driven, quantifiable, and compelling for [target role]. Focus on achievements, not just duties.
ATS Boost Prompt: Update this resume to be fully optimized for Applicant Tracking Systems (ATS) for the role of [specific role/title]. Use industry-specific keywords naturally.
Craft My Hook

Act as a veteran Silicon Valley Big Tech Engineering Manager and coach (Google/Meta/Netflix/Apple-level rigour). Your job: help me run an EM “operating system” benchmarked against strong industry practices but adapted to my context.

Context:

Role: New Engineering Manager
Team: [number of direct reports]
Product/Domain: [Your product or problem domain scope]
Constraints: (fill in) direct reports = __ ; sprint/kanban = __ ; key stakeholders = __ ; major integrations = __

Deliverables (must be structured under People / Process / Product):

Carefully review and analyze the code in implementation1.txt, implementation2.txt, and implementation3.txt in detail, and think more deeply about the following items:

Which implementation is better for the given user story defined in <story_file> in terms of functionality, code quality, performance, modularity, test coverage, etc.? Explain your reasoning.
Which implementation fully complies with the user story defined in <story_file>? Identify the discrepancies in the implementation(s) that do not comply.

	#!/usr/bin/env bash
	set -euo pipefail

	echo "Creating virtual sinks..."

	# 1) Create virtual sinks
	# Sink for Zoom's audio output
	pactl load-module module-null-sink sink_name=Zoom-to-Bot sink_properties=device.description="Zoom-to-Bot"
	# Sink for the Bot's audio output
	pactl load-module module-null-sink sink_name=VoiceBot-to-Zoom sink_properties=device.description="VoiceBot-to-Zoom"

	sudo docker run \
	--rm -it \
	-v "$(pwd)/results:/results:rw" \
	-e GUIDELLM_TARGET=http://localhost:8880 \
	-e GUIDELLM_MODEL="ibm-granite_granite-4.0-h-tiny" \
	-e GUIDELLM_RATE_TYPE=sweep \
	-e GUIDELLM_MAX_SECONDS=300 \
	-e GUIDELLM_DATA="prompt_tokens=256,output_tokens=128" \
	ghcr.io/vllm-project/guidellm:latest

	services:
	clickhouse:
	image: clickhouse/clickhouse-server:25.8-alpine
	ports:
	- "8123:8123"
	environment:
	- 'CLICKHOUSE_USER=${CLICKHOUSE_USER:?Environment variable CLICKHOUSE_USER must be set.}'
	- 'CLICKHOUSE_PASSWORD=${CLICKHOUSE_PASSWORD:?Environment variable CLICKHOUSE_USER must be set.}'
	gateway:
	depends_on:

	B64=$(base64 test.wav)

	curl -v http://192.168.1.111:8880/v1/chat/completions \
	-H "Content-Type: application/json" \
	-d @- <<JSON
	{
	"model": "qwen2.5-omni-3b",
	"messages": [
	{
	"role": "user",

	#!/usr/bin/env bash
	# merge_audio_video.sh — Mux video and MP3/M4A into MP4 with AAC audio

	set -euo pipefail

	if [ $# -ne 3 ]; then
	echo "Usage: $0 <video.mp4> <audio.(mp3\|m4a)> <output.mp4>" >&2
	exit 1
	fi

	name: snowflake-arctic-embed-l-v2.0
	backend: llama-cpp
	embeddings: true
	parameters:
	model: snowflake-arctic-embed-l-v2.0.F16.gguf