Ghoshan (@ghostyfreak) header
Ghoshan (@ghostyfreak) profile picture

Leading @TaxxaAI for the battle of accounting AI

Colombo
Joined November 2024
Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)Photo by Ghoshan (@ghostyfreak)
AI as your co-founder! Heard real frameworks and agentic workflows from idea to scale. Clear and practical. @Sithira269 @ghostyfreak @kevinmusprett @tharakahettihamu Hosted by @hatchonline . #gew2026
2
2
75
Oh boy, I'm not finding any reason to use gpt now. And thats really bad. Imagine @AnthropicAI won. no open model. no openai. Super intelligence in the hands of the very few.
4
497
Jensen might be the best real life rendition of Icarus
3
264
Mullaitivu grew faster than any district in Sri Lanka since 2012: +33%. In 2019 it also had the highest poverty rate: 44.5%. Asked in Claude, answered from census tables through our MCP.
1
2
218
Every MCP tool result should tell the model what to call next. ◇ `next`: ready-to-send calls ◇ `why` in plain words ◇ `matched_on` says when a synonym matched Fewer dead ends, fewer made-up field names.
1
169
Every data platform should ship an MCP before it ships a dashboard @developcountry
3
122
Type "power cuts". Get "Electricity supply reliability". ◇ One search parser for the site and the MCP ◇ Plurals stemmed, filler words dropped ◇ Synonyms always rank below typed words Nobody searches in government table titles.
2
136
This is how I measure what every MCP tool call costs. ◇ Proxy over D1, R2 and assets ◇ One metered env per call ◇ One Analytics Engine point per call
107
Attention is all you need applies to human more than agents
1
79
GOATED website anoma.ly/hex/ @opencode does everything, except for creating a competent harness. I am here for it!!!
1
100
Everyone’s at the mercy of their barber
2
73
Deepseek flash 1017 is a really good model for our benchmark. Its way cheaper and cites accurately. Now only if someone could make a token efficient version of this model?
82
Idk what’s wrong with me. I prefer how open ai models talk. Miles better.
I hate talking to OpenAI models now. Opus 5.5 spoiled me.
81
Models keep absorbing the harness. Half your tool clarifications and “remember to…” rules are now dead weight. Ablate them: leave-one-out on a real eval set, delete what doesn’t move the score. Your agent outgrew its harness. Audit for redundant instructions Harness²’s probe-and-compare loop is a good pattern, but point it at deletions and grade against real labels.
2
78
@bot can you do this for for our harness
1
54
Ours too. Something is happening
here are our teams commits since january something is happening
51
The fall of @conductor_build needs to be studied. If they had just open sourced, bet on the community, ship at the pace of progress of the community.
1
63
Im going ahead with @bunjavascript check for the 2x speedup. Its the fastest. Already part of the toolchain. @voidzerodev's tsrs complains about circular type errors
1
121