Open to AI/ML Engineer roles, internships, and freelance work
Deepanshu Dariya
AI / ML Engineer building production AI systems — from research and evaluation to deployed, observable services.
Multi-agent platforms, retrieval systems, computer vision, and MLOps. Every project ships with a real evaluation harness, honest metrics, and a path to production.
- Agentic AI
- LLMs & RAG
- Computer Vision
- MLOps
- Medical AI
0+
Projects shipped end to end
0
Domains: agents, CV, NLP, MLOps, medical AI
0+
Systems with evaluation harnesses
0
Years building with ML
Selected work
Systems built end to end — and measured honestly.
Each project below has a real evaluation harness, documented metrics, and a deployment story. Click through for the full engineering case study.
Trajectory
From computer-science foundations to agentic systems.
Foundations
Computer science & mathematics
Algorithms, systems, linear algebra, probability — the groundwork.
Deep learning
Neural networks, PyTorch, training pipelines
Reproducible experiments, honest baselines, disciplined evaluation.
Computer vision
Detection, segmentation, video understanding
Vision transformers for action recognition and spatio-temporal localisation.
LLMs & retrieval
RAG, fine-tuning, evaluation
Grounded question answering with citations, verifiable numeric answers, LoRA/SFT/DPO.
Reliability
Medical AI & model drift
Detecting, explaining, and mitigating distribution shift after deployment.
Now
Agentic systems & platform engineering
Multi-agent platforms, safe MCP tooling, cost/latency observability, human-in-the-loop.
Toolkit
What I build with.
Languages
Python · TypeScript · JavaScript · SQL · Java · C++
Machine Learning & Deep Learning
PyTorch · TensorFlow · Transformers · scikit-learn · LightGBM · CatBoost · Optuna · SHAP
LLMs, RAG & Agents
LangChain · LangGraph · MCP · RAG · vLLM · LoRA / SFT / DPO · FAISS · BM25 · Anthropic & OpenAI APIs · Qwen
Computer Vision
VideoMAE · TimeSFormer · ViT · DETR · OpenCV · YOLO
MLOps & Data Engineering
MLflow · DVC · Feast · dbt · DuckDB · Polars · Docker · Kubernetes · GitHub Actions · Kafka · Redis
Backend & Serving
FastAPI · Flask · Streamlit · PostgreSQL / pgvector · Neo4j · SQLite · REST · n8n
Hiring for AI/ML? Let's talk.
I'm open to full-time roles, internships, and freelance work in AI engineering, applied ML, and LLM systems.