Cognitive memory engine for AI agents. Sub-millisecond recall using Sparse Distributed Representations (SDR) instead of embeddings — no LLM calls, no cloud, no external dependencies. Pure Rust co...
Allow local LLMs to use Anthropic Skills, privately and securely
Perseus Vault is a single ~8MB Rust binary that gives AI agents durable, portable memory across sessions with no Docker, no Postgres, and no cloud. It is MCP-native (55 tools), local-first, and ful...
LLM4S is an open-source Scala-native framework for building production-grade LLM applications. Most LLM tooling today is Python-first, while many production systems in finance, telecom, data platfo...
A-Agent is an AI-powered workspace system that enables users to build specialized agent chains for complex tasks, transcending the limitations of single, monolithic AI models. Unlike traditional AI...
OBaI is an open-source agentic quant system that lets anyone research markets, test strategies, and run trading workflows with simple natural language. It makes advanced financial research more acc...
Neuroscience-inspired memory system for AI agents. Unlike passive vector stores, MemForge treats memory quality as something that should actively improve over time. It runs a 10-phase 'sleep cycle'...
Agent Memory Intelligence Benchmark. Existing benchmarks reward flashy retrieval metrics (Recall@k) that don't correlate with downstream task quality. Engram measures what actually matters: does th...
Modular open-source multi-agent AI and legacy hardware migration toolkit.
A dependency-free, model-neutral evaluator for wrong-subject attribution, information-cutoff violations, duplicate-evidence weighting, stale state, unresolved-state loss, false completion, improper...
An autonomous software-delivery harness: a task prompt becomes a planned, implemented, reviewed and pushed branch, driven by a fleet of single-purpose agents through plan, implement, review, intera...
An open-source, local-first, provider-neutral agent gateway with transparent tool calls, provenance and replay records, checkpointing and human review gates, so independent users can run and audit ...
Oleksandr Tepliuk
InstaVM
LLM4S Foundation