The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Deployed a multi-model CPU inference cluster on 2× HP ProLiant
DL360 Gen9 servers (376GB RAM total). Running 5 concurrent LLM
endpoints: Qwen3-Coder-30B, Qwen3-Next-80B, GLM-4.7-Flash,
Granite-4.0-Tiny — all quantized (GGUF/Q4_K_M). Optimized with
ik_llama + MKL for +62% throughput. Ollama-compatible API,
Open WebUI frontend, Opik observability.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Power your business with future-ready AI. ⚡
A goal-tracking sub-system that grounds an AI assistant in the user's full life-state — target amount, hard deadline, days remaining, current balance, pace status (ahead / on-pace / behind), and reflection streak.
Core features:
Multi-entry timestamped reflection log — plain Enter submits, Shift+Enter newline, optimistic UI, all entries inside one unified container with centered timestamp dividers
Ephemeral Summarize chat modal — loads today's braindumps as live AI context, brother-tone system prompt, conversation NOT persisted (privacy by design)
Nightly auto-summary cron at 23:55 — pulls reflections, runs through Claude, creates a standalone daily summary that downstream AI engines pick up automatically
Pace analysis — surfaces required-vs-actual daily rate, projected gap at deadline, status flag
Each AI engine pulling this data adapts its suggestions based on whether the user is ahead or behind on their goal — recursive context that compounds over time.
Tech: Claude API, SQLite, Node.js cron, custom retrieval logic.
A weekly and monthly AI suggestion engine that ingests the full user state and returns three ranked, goal-aligned recommendations per call.
What goes into the LLM context:
Chief Definite AIM (text, target, deadline, days remaining, current balance, pace status)
Perpetual goals (with AIM-link flags so aligned goals weight heavier)
Non-negotiable habits and their streaks
Recent insights with integration status (active / implemented / integrated:date)
Last 7–14 daily summaries
Past 28-day (weekly) or 90-day (monthly) suggestion history — for recursive learning
What it returns: three suggestion cards per call, each with priority + difficulty + AIM-alignment + tags + rationale.
Includes:
Intent gate — default AIM-aligned mode vs. custom prompt mode
Auto-detected response shape (ranked / analysis / hybrid)
Unified rate limiting across four endpoints (weekly default, monthly default, weekly custom, monthly custom)
Persistence to suggestion_history table for the next round's recursive context
Tech: Claude API, OpenAI API, custom prompt engineering, SQLite, Node.js, in-memory rate limiting.
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
portfolioreview
The best portfolios tell a story, not just show a grid. Share yours for feedback.
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?