class AgenticAIEngineer:
def __init__(self):
self.name = "Akash Shanmuganathan"
self.role = "Senior Agentic AI Engineer"
self.experience = "4+ years"
self.location = "Fort Mill, SC → Open to Relocation"
@property
def core_capabilities(self):
return {
"autonomous_systems": ["Multi-Agent Orchestration", "Voice AI", "Tool Integration"],
"llm_expertise": ["Gemini 2.0", "Claude 3.x", "Llama", "Mistral"],
"production_impact": ["99% intent accuracy", "75% latency reduction", "98% faster analytics"],
"specialization": "Building AI that thinks, decides, and executes autonomously"
}
def current_mission(self):
return "Architecting production agentic systems at Onedata (AWS Advanced Partner)">> CLEARANCE LEVEL: AWS Certified Data Engineer | 500K+ daily records processed | 99.7% SLA
🏦 NOVA BANKING AI [AUTONOMOUS FINANCIAL AGENT]
STATUS: ████████████░░░░░░░░ 60% COMPLETE
Voice-first banking agent that executes transactions, detects fraud, and solves issues autonomously
- 🎙️ Real-time voice interface with <200ms latency (Full-duplex, barge-in capable)
- 🧠 Gemini 2.0 Flash + Model Context Protocol (MCP) for tool orchestration
- 💳 7 autonomous banking functions: Balance checks, transfers, fraud detection, cashflow prediction
- ⚡ 99.8% transaction success rate | 8% escalation vs industry 80%
TECH: FastAPI WebSocket Vertex AI BigQuery React Web Speech API
🌐 CYBERPUNK PORTFOLIO [VOICE-CONTROLLED NEURAL INTERFACE]
STATUS: ████████████████████ 100% LIVE
Interactive AI-powered resume with JARVIS-inspired voice assistant - Not just a website
- 🗣️ Ask questions about my experience using voice (Web Speech API + Gemini)
- 🎮 Cyberpunk 2077 aesthetic with 3D environments (Three.js warp tunnels)
- 🧠 Context-aware AI that knows my entire professional history
- ⚡ 5,000+ visitors | 3.5min avg session vs industry 30sec
TECH: React 18 Three.js Gemini Pro Framer Motion TailwindCSS Express.js
🏗️ AGENTIC DATA ENGINEER [SELF-HEALING PIPELINE]
STATUS: ████████░░░░░░░░░░░░ 40% COMPLETE
3-agent system that profiles, monitors, and auto-fixes data quality issues 24/7
- 🤖 Profiler → Quality → Remediation agents with LLM reasoning
- ⚡ 2,500 rows/sec | 5x faster than Pandas | 84% memory reduction
- 🎯 98% quality score achieved autonomously from 73% baseline
- 🔧 Handles 23+ issue types with zero human intervention
TECH: Python LLMs DuckDB Polars Prefect Streamlit Great Expectations
🎯 INTELLIGENT ATS [SEMANTIC HIRING SYSTEM]
STATUS: ████████████████████ 100% PRODUCTION-READY
5-agent recruitment system using semantic search and LLM analysis - Understands context, not keywords
- 🧠 Deep semantic matching with vector embeddings (Qdrant)
- 🔍 94% correlation with human reviewers | Privacy-first local LLMs
- ⚡ 20-30s per candidate | Batch processing 50+ candidates/min
- 🎯 Specialized for Agentic AI and LLM engineering roles
TECH: FastAPI LangChain Ollama Qdrant Llama 3.1 Mistral Docker
💬 ASK MY CODE [CONVERSATIONAL CODE SEARCH]
STATUS: ████████████████████ 100% INDEXED
Chat with your codebase using natural language - RAG-powered semantic code search
- 💬 Natural language queries: "Show me all database connection logic"
- 🔍 99% recall accuracy | <1s response time for 100K+ docs
- 🧠 LLM explains code in business terms with context retention
- 🌐 Multi-language support: Python, JavaScript, Java, C++, Go, Rust
TECH: LangChain Mistral AI CodeLlama ChromaDB FAISS Streamlit Ollama
Agentic AI Engineer | Sep 2025 – Present | Fort Mill, SC
+ Built 4-agent platform with MCP → 75% latency reduction (8s → 2s)
+ Voice analytics with Amazon Nova Sonic → <300ms latency, 50+ concurrent sessions
+ Data lake processing 500K+ daily records → 99.7% SLA
+ Semantic search system → 99% recall, 94% response accuracy
+ NLP dashboard generation → 98% performance improvement (15min → 30s)Data Engineer | Jan 2021 – Jul 2023 | Chennai, India
+ Migrated 200 SQL workloads to AWS → 30% cost reduction
+ Real-time streaming pipeline → 10M records/day with Kafka + Spark
+ Query optimization → 45% performance improvement
+ Automated validation → 60% incident reduction| METRIC | IMPACT | SYSTEM |
|---|---|---|
| 🎯 Intent Accuracy | 99% | Multi-Agent Platform |
| ⚡ Latency Reduction | 75% ↓ (8s → 2s) | LLM Orchestration |
| 🚀 Dashboard Speed | 98% ↓ (15min → 30s) | NLP Analytics |
| 🛡️ Transaction Success | 99.8% | Voice Banking Agent |
| 💰 Cost Optimization | 30% ↓ | Cloud Migration |
| 🔍 Search Accuracy | 99% recall | RAG Systems |
| 📈 Data Processing | 10M records/day | Real-time Streaming |
| 🎙️ Voice Latency | <200ms | Full-duplex Voice AI |
Education:
M.S. Engineering Management:
- Northeastern University (2023-2025)
- GPA: 3.75/4.0
- Focus: Data-Driven Decision Making & AI Strategy
B.E. Electrical & Electronics:
- Anna University (2017-2021)
- Specialization: Signal Processing
Certifications:
- AWS Certified Data Engineer – Associate ✅
- AWS Generative AI Developer – Professional 🔄 (In Progress)
- AWS Advanced Partner Organization ⭐