🤖

Expertise

AI & Automation

From "can AI do this?" to "how do I remove this repetitive step?"

How this became a real interest, not a buzzword

My entry point into AI work wasn't chatbots — it was video editing. I was manually clipping long recordings down to highlight reels, and the repetitive part (finding the good moments, cutting, syncing) was exactly the kind of task a pipeline could do better than a person doing it for the tenth time.

That led to combining Whisper for transcription with an LLM for identifying interesting segments, then FFmpeg to actually cut the footage automatically. None of these pieces are exotic on their own — the value was in wiring them together into something that ran without me babysitting each step.

Whisper
Transcription layer
FFmpeg
Automated cutting

An example workflow

Long recording → short highlight clips, with minimal manual work

🎙️

1. Transcribe

The raw recording goes through Whisper to produce a timestamped transcript — this is the layer everything downstream depends on being accurate.

🧠

2. Extract highlights

An LLM reads the transcript and flags timestamp ranges likely to work as standalone clips — this is where prompt design actually matters, since vague instructions return vague picks.

✂️

3. Cut automatically

FFmpeg takes the flagged ranges and renders individual clips, so the manual work left is a quick review pass instead of scrubbing through the entire source.

Prompt engineering, practically

A pattern I return to constantly

Vague instructions produce vague output

"Find the good parts" gives an LLM nothing to grade against. Giving it a concrete definition — a joke landing, a clear answer to a question, a moment where tone shifts — gives it something to actually check for, and the difference in output quality is large.

Given this transcript with timestamps, return JSON:
[{ "start": "00:04:12", "end": "00:04:48", "reason": "..." }]

Only include a segment if it stands alone without
needing earlier context to make sense.
JSON out
Structured, parseable results
Clear rules
Not vague instructions

Lessons learned

🔁

Automate the boring 80%, not the judgment call

Full automation without a review step tends to ship mistakes confidently. I keep a lightweight human check before anything goes out publicly.

📐

Structured output saves more time than clever prompts

Asking for JSON with a fixed schema turned "parsing AI text output" from a fragile regex problem into a one-line JSON.parse.

🧭

The mindset shift mattered more than any single tool

Moving from "can AI do this" to "what repetitive step is actually costing me time" changed which projects were worth building at all.

Have a repetitive workflow worth automating?

Let's talk about what a pipeline for it could look like.

Get In Touch