Skip to content
View rehan243's full-sized avatar
🎯
Focusing
🎯
Focusing
112 followers · 495 following

Block or report rehan243

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
rehan243/README.md

waving hand    Typing SVG

GitHub followers  LinkedIn  Kaggle  profile visitors

divider

about me

i'm an AI/ML engineer based in the US. right now i'm building production AI systems at Reallytics.ai and Verticiti, mostly getting large language models to do useful things in the real world. not demos, actual systems with real users and real traffic.

before this i was at Afiniti and Cloud Kinetics for a few years. fraud detection, voice analytics, enterprise search. the kind of stuff that pages you at 3am when something breaks.

honestly what keeps me going is when an agent you built solves something you never explicitly told it to do. that feeling never gets old.

what i'm working on right now:

  • multi-agent systems that don't fall apart when you chain them
  • RAG pipelines that actually return relevant results
  • writing about what i learn every day, check it out here
 coding

divider

featured projects  rocket

Agentic AI Workflows
8 specialized AI agents with LangChain + OpenAI function calling. multi-agent orchestration with planning loops and guardrails. the project i'm most excited about right now.

RAG Enterprise Search
production retrieval pipeline over 2TB+ data. hybrid dense+sparse search with FAISS and BM25, cross-encoder re-ranking. deployed on AWS SageMaker.

Voice AI Platform
real-time voice infrastructure handling 500+ concurrent calls. WebSockets, Kafka, VAD, streaming STT. built the sentiment analysis piece from scratch.

LLM Fine-Tuning LoRA
fine-tuning LLaMA and Mistral with LoRA/QLoRA/PEFT. 40% cheaper than hosted APIs. includes the full training loop, data pipeline, merge + quantize scripts.

RLHF LLM Optimization
full RLHF pipeline: reward model with Bradley-Terry loss, PPO trainer with KL scheduling, DPO as an alternative. 68% win rate on eval, 96% safety compliance.

Sentinel Fraud Detection
ensemble XGBoost + neural net with 650+ engineered features. Redis-backed real-time velocity scoring, SHAP explainability, Kafka alert routing.

view all repositories

divider

tech stack

not going to pretend i use everything equally. here's what i actually reach for:

tech stack

the full picture (click to expand)
daily drivers Python, PyTorch, FastAPI, Docker, Git, VS Code
LLM and GenAI LangChain, LlamaIndex, HuggingFace Transformers, vLLM, PEFT/LoRA/QLoRA
data and vector FAISS, ChromaDB, Pinecone, PostgreSQL, MongoDB, Redis, Kafka, Elasticsearch
cloud and MLOps AWS (SageMaker, Bedrock, Lambda, ECS), GCP Vertex AI, Azure OpenAI
ML frameworks TensorFlow, scikit-learn, XGBoost, LightGBM, ONNX
infrastructure Kubernetes, Terraform, GitHub Actions, MLflow, Weights & Biases

divider

github stats

github stats streak stats

top languages

trophies

trophies

contribution graph

contribution graph

my github contributions eating themselves

contribution snake

divider

recent writeups

i write about what i'm building and learning. nothing polished, more like notes to my future self that happen to be public.

Retrieval Augmented Generation Rag In Production

Retrieval Augmented Generation Rag In Production
2026-06-13

 Large Language Model Llm Fine Tuning With Parame

Large Language Model Llm Fine Tuning With Parame
2026-06-13
Real World Applications Of Reinforcement Learning

Real World Applications Of Reinforcement Learning
2026-06-10

 Real Time Data Streams For Ml Model Training

Real Time Data Streams For Ml Model Training
2026-06-09

📚 View all articles →

recent activity

💬 Commented on [Bug]: MiniMax-M3-MXFP8 hangs on FlashInfer MNNVL all-reduce in vllm-project/vllm (2026-06-16)

💬 Commented on ci: streamline GitHub Actions test coverage and runtime in feast-dev/feast (2026-06-16)

💬 Commented on [Bug]: Artifact integrity report follows security policy in wandb/wandb (2026-06-16)

💬 Commented on bug: async delete_all race condition corrupts entity store l in mem0ai/mem0 (2026-06-16)

💬 Commented on Community Case Study: Structured constraints improve code ge in zai-org/ChatGLM-6B (2026-06-16)

💬 Commented on [provider-mapping-sweep] cohere: prompt-cache hit count (usa in pydantic/pydantic-ai (2026-06-16)

💬 Commented on API-key (Bearer) auth still depends on Keycloak offline sess in OpenHands/OpenHands (2026-06-16)

💬 Commented on First-party Agent Skills support in langchain (without depen in langchain-ai/langchain (2026-06-16)

what i'm reading lately

stuff i've been digging into recently. mostly papers, blog posts, and rabbit holes that kept me up too late.

🔬 AI Model Monitoring, Evaluation, and Automated Feedback Loops

🔬 Retrieval-Augmented Generation (RAG) in Production Search and Chat Applications

🔬 Large Language Model (LLM) Fine-Tuning with Parameter-Efficient Techniques

🔬 Causal Inference and Discovery in Observational Data

🔬 Real-World Applications of Reinforcement Learning from Human Feedback

🔬 Real-Time Data Streams for ML Model Training

code snippets

📌 Streaming JSON Parser for Large Language Models — Production Pattern (Python) (2026-06-16)

📌 Streaming JSON Parser for Large Language Models — Production Pattern (Python) (2026-06-16)

📌 Multi-Provider LLM Router with Fallback — Production Pattern (Python) (2026-06-15)

🤖 Profile auto-updated on 2026-06-16 17:15 UTC

divider

if you made it this far, you should probably just say hi

connect on linkedin   follow on github

Pinned Loading

  1. Voice-AI-Platform Voice-AI-Platform Public

    Real-time voice AI infrastructure — 500+ concurrent calls, WebSockets, Apache Kafka, gRPC/C++ with CUDA. Speech-to-text, sentiment analysis, sales insights.

    Python

  2. Agentic-AI-Workflows Agentic-AI-Workflows Public

    Production AI Agents for enterprise automation — 8+ specialized agents using LangChain, OpenAI function calling, and FastAPI. Multi-agent orchestration, tool use, planning loops, guardrails.

    Python

  3. BiiView-Object-Detection BiiView-Object-Detection Public

    Real-time object detection with Meta AI Segment Anything Model (SAM) — 90% accuracy across 11M+ images and 1.1B+ segmentation masks.

    Python

  4. Digital-People-Platform Digital-People-Platform Public

    Hyper-realistic talking avatars — SadTalker lip-sync + Microsoft SpeechT5 TTS + OpenAI conversational AI. 70% realism improvement.

    Python

  5. LLM-Fine-Tuning-LoRA LLM-Fine-Tuning-LoRA Public

    Fine-tuning LLaMA-2, Mistral with LoRA, QLoRA, PEFT — 40% cost reduction vs hosted APIs. VLLM serving with CUDA optimization on AWS SageMaker.

    Python

  6. RAG-Enterprise-Search RAG-Enterprise-Search Public

    Production RAG pipeline — enterprise knowledge retrieval across 2TB+ data using LangChain, FAISS, ChromaDB, PG-Vector with cross-encoder re-ranking. Deployed on AWS SageMaker.

    Python