Pinned Loading

1. DiSpec DiSpec Public

   DiSpec — a from-scratch LLM inference engine: paged attention, continuous batching, CUDA-graph decode, speculative decoding, and prefill/decode disaggregation

   Python

2. Parallel-Eagle Parallel-Eagle Public

   A lossless speculative decoder: a parallel multi-token drafter + dynamic tree verification.

   Python

3. Schema-Linking-with-SFT Schema-Linking-with-SFT Public

   LoRA fine-tuned Qwen2.5-Coder-1.5B for schema linking in NL-to-SQL: given a question and a database schema, predicts the referenced tables and columns as JSON. Uses self-consistency decoding (k=10)…

   Python

4. search-hotstar search-hotstar Public

   Semantic search engine over a streaming catalog — scrapes, enriches, embeds, and indexes titles into Qdrant, then serves vector + lexical hybrid search via FastAPI with a lightweight frontend.

   Python

5. news-recommendation-system news-recommendation-system Public

   News recommendation

   Python

6. CSE291g-MIC-winter-2026/humerus_atlas_building CSE291g-MIC-winter-2026/humerus_atlas_building Public

   Repository containing the implementation for Humerus Atlas construction using Implicit Neural Representations (INRs), developed as part of the CSE291G course project (Winter 2026).

   Python