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
What if an AI assistant could tell you exactly which law supports its answer?
That's the idea behind Seera, an AI-powered legal research platform I'm building to make Ethiopian legal information easier to explore.
Instead of relying only on an LLM's general knowledge, I built a retrieval-based approach that searches indexed legal documents and grounds answers in relevant sources.
The stack includes:
Go for the backend and streaming responses
React + TypeScript for the frontend
PostgreSQL + pgvector for legal document retrieval
RAG to connect AI-generated answers with source material
One design principle matters to me: when the available sources don't support an answer, the system should make that clear rather than pretend to know.
Seera is still a work in progress, but building it has taught me a lot about combining backend engineering, AI, and a real-world problem.
Building useful things, one challenge at a time.
Feel free to check it here https://www.seera.site/
I'd test Seera with a question that retrieves a closely related law but asks for a conclusion the passage doesn't support. Does it still abstain when retrieval similarity is high?
Thanks for checking on it. Currently there only two law legal documents with 878 articles are cited and Seera only answers Grounded data.
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
CRAG — Self-Correcting Research Assistant
Overview
A research-assistant project exploring how AI applications can retrieve information and improve the quality of their responses through a self-correction workflow.
Project focus
Python-based AI application development.
Retrieval and research workflow design.
Improving answer reliability through evaluation and correction.
Technologies: Python, AI, retrieval-augmented generation.
Self correction is where RAG gets useful. How do you judge a retrieved chunk as bad enough to trigger a new search?
My AI assistant runs all day. The expensive model only does the thinking.
Most requests need no AI at all. "Call me in 30 minutes" is handled by plain code. Reminders, schedules and limits are code too.
Short tasks like sorting and scoring go to fast models on a free tier.
The big model only gets planning and long conversations, where it actually has to think.
If it hits a limit, a backup model answers instead of the app going silent.
If your AI app's bill keeps growing, check what really needs the big model.
What does your app send to the most expensive model?
Same pattern holds in outbound tooling. Filtering, dedupe and enrichment lookups are plain code. The model only writes the first line once the row is already verified. Cuts cost and the weird outputs.
I built an open-source skill that stops AI from making every design look the same
Every AI-generated website looks the same. Centered hero. Purple gradient. Three feature cards.
The reason isn't bad taste. AI agents fill gaps in a brief with the average of their training data. If you don't define what "not average" means for your project, you get the mean.
So I built Intent Interrogator, a free skill that works across Claude Code, Codex, Cursor, Copilot, and any agent that reads the Agent Skills standard.
It does three things:
Interrogates - asks 4 questions before writing any code:
→ What should someone feel in the first 3 seconds?
→ Who specifically is this for?
→ What's the one constraint I must respect?
→ What should this definitely not look like?
Persists - saves the brief to a .intent/ file inside your project. New chat, new week, new agent, the intent is still there.
Verifies - before declaring the work done, it compares what it built against what you asked for, and produces an Intent Receipt showing where it matched and where it drifted.
That last part is the piece I haven't seen anywhere else.
You don't need to code to use it. You just answer four questions.
Free forever. Feedback welcome.
#design #ai #opensource #productdesign
github.com
GitHub - yutivora/intent-interrogator
Contribute to yutivora/intent-interrogator development by creating an account on GitHub.