On AI Lounge since Nov 2026
Aagam Sachin Bothara
Member #4,891FounderL2AI inference, UIC
MS CS @ UIC | Cloud & Distributed Systems | LLM Inference • Performance • Cost Optimization | ServiceNow (ITSM/ITOM)
How karma works. Everyone starts with 5. You get +1 each time another member upvotes one of your posts or comments (your own upvotes don't count), and lose it if they take the upvote back. Post what you're building, answer questions and help other members to earn more.
No posts yet
Aagam Sachin Bothara hasn't posted yet.
Pinned projects
Joule is an execution runtime for AI agents that does not trust an agent just because it says “I’m done.” It verifies the work externally and only spends additional agents when the current result is actually failing.
ai-agents
ModelGuard is a practical ML monitoring and MLOps system that tracks data and prediction drift in deployed models. It explains when and why models degrade, assesses the severity of change, and supports human-approved retraining workflows to keep models reliable in production over time.
Retrieval-Augmented Generation (RAG) system with hybrid search (FAISS + BM25), cross-encoder reranking, and confidence-aware answer verification. Detects retrieval failures, performs multi-hop query decomposition, flags contradictions across sources, and safely abstains or requests clarification when confidence is low.
Measures which LLM serving settings pay off on your GPU and your traffic, then serves the winner. Benchmarks vLLM, SGLang and llama.cpp configurations; one OpenAI-compatible AP
About
MS CS @ UIC | Cloud & Distributed Systems | LLM Inference • Performance • Cost Optimization | ServiceNow (ITSM/ITOM)