Hello old friend!
🔸Urgent+Important = DO
🔸Urgent+Non-Important = DELEGATE/SYSTEMISE (Me: Automate!!)
🔸Non-Urgent+Important = PLAN
🔸Non-Urgent+Non-Important = AVOID
Panelist from Thailand, Non, shared how his last visit to NZ, 30 years ago, for a summer camp, changed his life trajectory. Got his life back to getting a PhD from Harvard.
Been wondering what's the link of this session with ALGIM and NZ, turns out there's a session before lunch on Christchurch as a Smart City. Michael Healy, Christchurch City Council was the presenter.
Can't help but notice that Singapore is missing on the panel.
Indonesia. Smart city is not just about technology, it's actually about the people. Building capability is a key focus.
Malaysia smart cities. By 2030, at least 93 city/district councils out of 160+, will achieve smart city level 3. https://www.malaysia.gov.my/portal/content/30946
Non-govt article on the different levels https://www.smartcityacademy.my/smart-city-initiatives/
Super excited about their next steps. Collaboration being the highlight. Bright future for auto-classification using LLM. 🔆 🌱🌼🌻
(4/4)
93% accuracy. Wild success. Est savings of 160 hours per annum (1 mth of FTE)
Azure dev and prod cost: ~$332 NZD
QA cost: ~$30 NZD
Actual production run: ~$150 NZD
💪💪🤩😮😮
Learnings: corruption detection, metadata enhancement, composite documents, summaries
(3/🧵)
#ALGIM2024 #ALGIM24
LLM for Classification (ChatGPT 3.5 Turbo). Lots of prompt engineering. Lots and lots of iteration to get the prompt right.
Trialled llama, gemini, on local GPUs, just couldn't get the accuracy they need.
Total cost depends on the total number of tokens. Discovered more is not always better.
QA key part of the detailed solution. Sampling. Might result is our first EDRMS disposal (Yayyy!)
(2/🧵)
joshuatj