Agentic AI Developer & Team Lead
Syed Abdul Kareem
I build production-grade autonomous AI systems — multi-agent orchestration, RAG pipelines, and brain-computer interfaces.
01 — About
Building AI that ships, not demos
I'm an Agentic AI Developer leading a team at IT Automation LLC (US, remote), where we ship autonomous agentic systems for legal, financial, and public-safety domains — LangGraph orchestration, RAG pipelines, and FastAPI services running in production.
Before that, I built real-time EEG acquisition and signal-processing systems at IHub-Data, IIIT-Hyderabad, working with OpenBCI hardware across SSVEP, Motor Imagery, and P300 neural paradigms — research currently under publication.
I care about AI systems that are observable, evaluated, and grounded: LLM-as-judge workflows, hallucination mitigation, and trace debugging are part of how I ship, not an afterthought.
- Production agentic systems shipped
- 0
- Best model accuracy (ResNet50)
- 0%
- Latency cut via parallel agents
- 0%
- Public repositories
- 0+
Production agentic systems shipped
Best model accuracy (ResNet50)
Latency cut via parallel agents
Public repositories
02 — Stack
Tools I think in
Agent orchestration first, classical ML underneath, and the infrastructure to run both in production.
AI & Agents
01- Agentic AI
- Multi-Agent Systems
- LangChain
- LangGraph
- RAG
- OpenAI
- Anthropic Claude
- Gemini
- LLaMA
- Groq
- Prompt Engineering
ML & Deep Learning
02- Deep Learning
- NLP
- Computer Vision
- MLOps
- Model Evaluation
- Signal Processing
- OCR
- Document Understanding
Frameworks
03- FastAPI
- Flask
- TensorFlow
- Keras
- Scikit-learn
- Streamlit
- Pandas
- NumPy
- OpenCV
- React
Data & Infra
04- FAISS
- ChromaDB
- MySQL
- Docker
- AWS
- GCP
- Git
- Linux
- CI/CD
Languages
05- Python (Advanced)
- SQL
- TypeScript
- Java
03 — Experience
Where I've built
IT Automation LLC
Feb 2026 — PresentAgentic AI Developer & Team Lead · United States · Remote
- Leading a team of 2 engineers building autonomous agentic AI systems across three production applications.
- Built an Agentic Legal Case Management System — LLaMA 7B case summarisation, context-aware RAG chatbot, strict RBAC.
- Developed an Agentic Loan Origination System automating underwriting: credit fetching, risk analysis, and memo drafting in under 1 second.
- Shipped a Service Request Management app with agentic case routing, Groq-powered chatbot, and duplicate detection for public-safety requests.
- Implemented RAG pipelines with vector databases, structured/unstructured extraction, and OCR ingestion across all three platforms.
- LangGraph
- RAG
- FastAPI
- Multi-Agent
IHub-Data, IIIT-Hyderabad
Aug 2025 — Jan 2026Software Developer — Neurotechnology · Hyderabad · Research under publication
- Developed a real-time EEG acquisition and signal-processing system on OpenBCI Cyton+Daisy supporting SSVEP, Motor Imagery, and P300 paradigms.
- Optimised low-latency inference pipelines and built brainwave visualisation tools with research engineers, advancing autonomous human-machine interaction.
- OpenBCI
- EEG
- Signal Processing
- Python
Innodata
OngoingLLM Evaluator & Data Annotator — Freelance · Remote
- LLM evaluation and AI data annotation, including work for Meta SuperIntelligence Labs.
- LLM Evaluation
- Meta SuperIntelligence Labs
04 — Work
Selected projects
Flagship builds with the receipts — problem, approach, and measured outcomes.
In production at IT Automation LLC — company confidential
Agentic Legal Case Management
LLaMA 7B summarisation · RAG chatbot · RBAC
Agentic Loan Origination
Automated underwriting in <1s · LangGraph agents
Service Request Management
Agentic routing · duplicate detection · Groq chatbot
05 — Contact
Let's build something autonomous
Open to full-time AI/ML roles, consulting, and research collaboration.