Skip to content
View hviana's full-sized avatar
108 followers · 146 following

Achievements

Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: Pull Sharkx2Achievement: Starstruckx2

Achievements

Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: Pull Sharkx2Achievement: Starstruckx2

Block or report hviana

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
hviana/README.md

Henrique Emanoel Viana

AI Engineer & Computer Scientist AI agents in production · LLM orchestration · reasoning architectures M.Sc. in AI and Multi-Agent Systems, UTFPR · ORCID

I build AI systems that run in production, and research architectures for machine reasoning.

Sema — retrieval and reasoning in one search

sema combines high-dimensional vector search with an explicit knowledge graph. It indexes embeddings through an IVF structure over 1-bit quantized codes, and answers by chaining stored facts — retrieval and inference as a single search.

Output is deterministic, auditable and traceable to source, on CPU: an alternative to conventional RAG where hallucination and lack of traceability are unacceptable.

vector search · knowledge graph · explainable AI · no GPU Paper · npm

botIA — multi-agent AI in production (2025–2026)

Cofounder and software architect. A WhatsApp multi-agent platform: agent orchestration, tool calling, human handoff, AI-based lead classification, and integrations with ERP, e-commerce, CRM and calendar systems. Ceased operations in 2026.

Research

M.Sc. on explainability in deliberative multi-agent systems — reconstructing agent decisions causally from internal-state transitions.

Python · TypeScript · LLM APIs · backend · AWS · Docker

LinkedIn · [email protected]

Pinned Loading

  1. faster faster Public

    A fast and optimized middleware server with an absurdly small amount of code (300 lines) built on top of native HTTP APIs with no dependencies. It also has a collection of useful middlewares: log f…

    TypeScript 57 11

  2. faster_react faster_react Public

    Tiny Full-Stack React framework. Avoid Overengineering. Automatic routes, reload and component bundle. It uses its own RSC engine, combining SSR and CSR. 100% Deno, no Node dependencies. Fully comp…

    TypeScript 99 5

  3. deno_kv_fs deno_kv_fs Public

    Deno KV file system, compatible with Deno deploy. Saves files in 64kb chunks. You can organize files into directories. You can control the KB/s rate for saving and reading files, rate limit, user s…

    TypeScript 17 1

  4. faster_react_core faster_react_core Public

    faster_react core files

    TypeScript 3

  5. sema sema Public

    Sema is a non-parametric, instance-based reasoning system: a Vector Symbolic Architecture over a content-addressable memory, with inference by weighted automated deduction.

    TypeScript 6