ABL-000 · SELF-HOSTED MIND · DESCENT VECTOR LOCKED
Custom agents, workflow automation, and self-hosted LLM deployment — shipped as working software, not slideware. From the model weights up to the browser automation that drives the whole thing.
LOG // 1 CRAFT INBOUND · PILOT: A. HOSSAIN · DESTINATION: LAB
29.9M
parameters pretrained from random initialisation — 2.0B tokens, 9.0 hours, one RTX 3070 Ti. Best val loss 1.341. Own architecture, own tokenizer, own weights.
27B
served on that same 8 GB card — CUDA llama.cpp from source, GGUF quantization, MTP self-speculative decoding. Measured, not guessed →
128
tests gate every deploy of the platform. 7,600+ records in the store, 450+ commits, one operator running it under pm2.
36/0/0+31
agents built / partial / scaffold, graded from source by a published rule → — 300+ lines and no stub markers, not self-assessment. The rule is on the page so the count can be checked. The +31 are specified and not built — each with the files it would touch and how it fails safe — counted separately, because a specification is not an agent.
REC 01 // MISSION RECORD
A production personal-AI system coordinating many specialized agents behind one API. FastAPI backend routing across a fleet of locally run models, a vector-searchable store I schema-designed, and an egress boundary that classifies every outbound request and keeps sensitive data on the machine.
REC 02 // MISSION RECORD
Separated analytics, data, decision, execution, simulation, dashboard, and testing layers. Sole-authored the React 19 + TypeScript operator dashboard, plus SQL-backed ingestion and evaluation pipelines with logging and reporting.
REC 03 // MISSION RECORD
Running 27B–35B models on a single 8 GB consumer GPU: CUDA llama.cpp built from source, GGUF quantization, MTP self-speculative decoding, and MoE expert-offload tuning. Measured, not guessed.
REC 04 // MISSION RECORD
Discord bots on live servers: 80+ registered operations across 20+ modules, persistent per-user state, automated moderation, and an authorization layer with permission tiers that gates destructive operations behind explicit approval.
Git throughout. Readings on the right are live where the lab can actually see them.
29.9M parameters, my own architecture code and my own tokenizer, random weights to coherent prose in sixteen minutes — nine hours on one RTX 3070 Ti, including the two silent bugs that made the first attempt eighty-six times too slow.
TX 04.1Six experiments on a single RTX 3070 Ti. Multi-token prediction doubled throughput for free, speculative decoding made things worse, and the biggest win was a Windows setting silently moving VRAM into system RAM.
TOUCHDOWN · ALT 0 M — CRAFT SECURED · PILOT HOME
■ PRESSURE EQUALISED · ENTERING
Nobody let me in. Nobody was here. The lanes had been running the whole time I was away.
Tell me the problem in a sentence or two. If it's a fit I'll come back with a scope — usually same day.
Not here to hire? Everything else — TikTok, Discord, Hugging Face, and who I actually am — is on about & signals.
THIS PAGE · MEASURED IN YOUR BROWSER
Hand-written HTML, CSS and JavaScript — no framework, no build step, no component library. These are read from your browser when the page finishes loading, so they are yours, not mine: reload and they move with your connection and your cache.