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
I was avid TTRPG DM and Player, this means I have a plenty of rulebooks from different systems. Most of them in PDF and they are horrible in size. I don't want to use online services and most of services have a limits on size or quantity of pdf for processing.
"Scratch my own itch" approach leads to create pdfslim. Offline tool for pdf compression and optimization. Allows bulk-processing, memory tweaking and sufficient level of compression (HtR from 145 Mb to 10 Mb for example).
github.com
alexander-yatskov/pdfslim
Offline tool for pdf compression and optimisation. Contribute to alexander-yatskov/pdfslim development by creating an account on GitHub.
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
I built a local, synthetic demo of Theme Handoff Pack for agencies that repeatedly hand off small theme changes.
You supply an explicit file list, existing check notes and rollback instructions. One Python command assembles linked HTML, a SHA256 file manifest and a ZIP. It packages what you recorded; it does not test Shopify, validate a repair or establish client acceptance.
The current prototype accepts synthetic samples only. Its CSS change and drawings are fictional, not client work or browser screenshots. Unrun checks stay visibly unrun. The tool has no network calls or AI runtime; Codex produced the code and documentation, with the individual operating ThemeCare accountable.
I am testing a USD29 one-time price for a future reusable local version. No payment or booking is accepted now.
For agencies: thinking about your last handoff, would this replace a step in your Git/Markdown/ZIP process? Would you actually buy it at that price? What required input or command-line setup would make it slower?
Please describe your workflow in general terms. Keep private client files, names and credentials out of public replies. Compliments and views will not be counted as purchase evidence.
This week I worked on a widget library for a new SaaS product.
On complex products, a widget library can pass 100 widgets fast. Every one needs multiple sizes, empty and error states, permission rules, and has to handle real data that never looks like the mockup. Without a system, each new widget becomes a one off and the product slowly starts to feel stitched together.
Enterprise RAG AI: Private Doc Intelligence
A fast-growing B2B software firm needed a secure, enterprise-grade Retrieval-Augmented Generation (RAG) system to query thousands of confidential internal documents (legal contracts, technical architecture specs, Notion pages, and employee SOPs) in real time without third-party data leakage.
Key Challenges:
High hallucination risk and irrelevant context retrieval using standard keyword search.
Managing multi-format files (scanned PDFs, tables, markdown) with strict access control based on user hierarchy.
Technical Implementation & Solution:
• Engineered an automated ingestion pipeline in Python with LlamaIndex and LangChain for semantic chunking and metadata preservation.
• Integrated a hybrid vector search architecture pairing dense embeddings (OpenAI text-embedding-3) with sparse BM25 indexing in Qdrant.
• Added a cross-encoder re-ranking layer (Cohere Re-ranker) to prune low-relevance chunks, reducing API token costs by 40%.
• Built robust Role-Based Access Control (RBAC) via FastAPI endpoints, ensuring departmental data isolation.
Results & Impact:
• Dropped internal information retrieval time from 20 minutes to under 4 seconds.
• Attained 98% factual precision with traceable in-line page citations.
• Full on-prem Docker deployment ensuring 100% private data security.
