Skip to main content
Back to Projects
Autonomous AI AgentsMar 2025 → Present

Agentic Performance Appraiser

Multi-agent evaluation framework that scores code changes by combining SonarQube quality metrics with Git activity signals, then surfaces the verdict to engineering managers.

PythonLangGraphSonarQube APIGit APIFastAPIPostgreSQL
Signals fused
Quality + Activity
Review prep time
−60%
Verdict latency
<10s

Problem & Solution

The Problem

Performance reviews for engineers are slow and inconsistent. Quality and code-activity signals exist in separate systems (SonarQube, Git) but never get combined into a single, fair score.

The Solution

Built a multi-agent framework that pulls per-developer signals from SonarQube (code quality) and the Git API (commit activity, review participation), fuses them through a planning agent, and produces a structured performance verdict for engineering managers.

System Architecture

End-to-end flow from intake to outcome

  1. 01

    Quality Collector

    SonarQube API

    Agent pulls per-developer SonarQube quality metrics over the review window.

  2. 02

    Activity Collector

    Git API

    Agent pulls commit and review activity from the Git API.

  3. 03

    Fusion Agent

    LangGraph

    LangGraph orchestrator fuses the two signal streams into a structured verdict.

  4. 04

    Verdict Service

    FastAPIPostgreSQL

    FastAPI exposes the verdict to the manager UI and stores it for audit.

Engineering Challenges & Wins

Optimization results and engineering tradeoffs

Optimization Results

  • Multi-agent design with separate collectors for quality and activity signals.
  • SonarQube + Git API integrations via a LangGraph orchestrator.
  • Structured verdicts exported to engineering managers via FastAPI.

Engineering Challenges

  • Combining qualitative and quantitative signals without biasing the verdict.
  • Keeping the agent's reasoning auditable for HR review.

More case studies

Explore other production systems I've engineered end-to-end.

All Projects

Related Projects

Autonomous AI AgentsFeatured
OmniShield ICES: Asynchronous AI Email Security Middleware

Mail-server gateway (Postfix Milter) that intercepts every inbound message, runs a fine-tuned BERT core (M-BSCE) through an asynchronous inference pipeline (RabbitMQ + Celery), and adds multimodal threat detection (PaddleOCR for Quishing) with SOAP clawbacks to the upstream mail system.

Quishing detection
98%+
p95 classification
<400ms
Async throughput
5k msg/min
PythonPostfix MilterRabbitMQCeleryBERT+6
View Case Study
Autonomous AI AgentsFeatured
OpenClaw Multi-Channel Orchestration

Autonomous multi-channel AI platform that unifies chat, voice, and email behind microservices and an API gateway, with shared context across channels.

Channels
3 unified
Tenant onboard
<1 day
Cross-channel memory
100%
OpenClawFastAPIWebRTCWebSocketRabbitMQ+4
View Case Study