Skip to content
← All projects

AI/ML · Data Engineering · Full Stack

NEXUS AI

Multi-agent research and execution workspace

Status
Complete · actively extended
Year
2026
Live application
Runs locally with Docker Compose and Ollama. Not publicly deployed.
Mission Control showing a multi-agent mission timeline
citation grounding checks
6/6
citation grounding checks
unsafe SQL statements blocked
8/8
unsafe SQL statements blocked
canary leaks
0/5
canary leaks

01 Problem

Most AI apps return one opaque answer. NEXUS persists everything that produced it: the plan, the agents that ran, every tool call, the cited passage, the query behind each figure and a verification report - so a result can be inspected, replayed and exported.

02 Product overview

Missions run as a LangGraph graph: a planner proposes steps, a human approves, research, document and data agents execute, then a synthesizer, an evidence verifier and an artifact generator finish the job. Missions can pause, resume, cancel and retry from a checkpoint. The workspace also includes a knowledge library with hybrid retrieval, a DuckDB data lab, an artifact studio, a workflow builder and an authenticated MCP server.

Implemented

  • Streaming chat workspace with four modes
  • Mission Control: Planner → Human approval → Research / Document / Data agents → Synthesizer → Evidence verifier → Artifact generator
  • Pause, resume, cancel and retry-from-checkpoint via a Postgres checkpointer
  • Knowledge library with pgvector + full-text retrieval fused by RRF, plus a PDF viewer
  • Data Lab with a DuckDB sandbox, guarded SQL and typed operations
  • Authenticated MCP server over Streamable HTTP

Planned, not yet built

  • Consuming external MCP servers
  • OCR
  • NLI-based verification
  • Reranker
  • Multi-worker scaling
  • OpenTelemetry

Credit: Evaluation quality gates were contributed by collaborator rithika-sr (merged PR #1).

03 System architecture

SSENext.js UIworkspace · missionsMCP clientbearer tokenFastAPIREST · SSEarq workerLangGraph missionsPostgrespgvector · checkpointsRedisstreams · queueOllamaqwen3 · embeddingsSearXNGweb researchDuckDBdata lab sandbox
API and worker share Postgres (vectors, checkpoints) and Redis (chat streams, task queue). Models run locally through Ollama.clientservicestoremodelqueue

04 Core technologies

Agents & AI

  • LangGraph
  • Ollama (qwen3:4b-instruct)
  • nomic-embed-text
  • MCP
  • SearXNG

Data

  • PostgreSQL + pgvector
  • DuckDB
  • sqlglot
  • Redis
  • arq

Application

  • FastAPI
  • Next.js 16
  • React 19
  • Monaco
  • Radix

05 Engineering challenges

Small models, structured output

A 4B local model doesn't reliably emit valid plans. Constrained decoding plus a single repair retry made outputs dependable without a bigger model.

Plans that stay on the rails

Instead of free-form plans, the analyst agent selects from a derived catalog of operations, which keeps execution safe and testable.

Durable human approval

Missions pause for approval and must survive restarts. A LangGraph Postgres checkpointer makes interrupts durable.

06 Technical decisions

  1. 01Deterministic evidence verifier instead of an LLM judge
  2. 02Redis streams for chat; Postgres plus pub/sub for tasks
  3. 03Self-hosted SearXNG for web research
  4. 04No arbitrary code execution anywhere in the agent loop

07 Verified test results

Backend
146
Frontend (Vitest)
20
Offline
49
Playwright journey
1

Latest GitHub Actions run on main: passing. Counts as reported in the repository's README and testing docs.

08 Screenshots

Mission Control showing a multi-agent mission timeline
Mission Control: plan, approval and agent timeline
Citation inspector highlighting a source passage
Citation inspector linking claims to passages
Data Lab with a generated chart over DuckDB data
Data Lab: guarded SQL over DuckDB
Streaming conversation workspace
Streaming workspace conversation

Screenshots captured from the running application and published in the repository's docs.

09 Links

No public deployment. Runs locally with Docker Compose and Ollama. Not publicly deployed.