Skip to content
View QHarshil's full-sized avatar
7 followers · 6 following

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx2

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx2

Highlights

  • Pro

Block or report QHarshil

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
QHarshil/README.md

Harshil

Applied AI · ML Systems · Backend · Search · Infrastructure

AI applications, inference & model serving, search/ranking, and backend systems.

Python C++ Go Java TypeScript SQL

Open source

Contributing to:

vLLM · GuideLLM · OpenSearch k-NN · Kubernetes · EF Core · PyTorch

Recent work includes inference correctness, benchmarking, vector search, Kubernetes API performance, and query translation.

Building

Project What it is
CineMatch Grounded AI recommendation agent over hybrid retrieval and learned ranking
Anytime Inference Planner Deadline-aware ML serving with continuous batching and adaptive admission
SemanticSearch Hybrid semantic + keyword retrieval with filtering, caching, and ranking evaluation
Execution Copilot Market execution simulator for testing execution strategies under different latency and venue conditions
Stack

Languages: Python, C++, Go, Java, TypeScript, SQL
AI / ML: PyTorch, ONNX Runtime, Transformers, RAG, tool calling, LightGBM
Data / Search: PostgreSQL, pgvector, OpenSearch, Elasticsearch, Redis, HNSW
Systems: AWS, GCP, Docker, Kubernetes, gRPC, Kafka

Pinned Loading

  1. Anytime-Inference-ML-Model-Serving Anytime-Inference-ML-Model-Serving Public

    Deadline-aware configuration selection for ML inference systems.

    Python

  2. CineMatch CineMatch Public

    A film and TV recommender with a grounded AI assistant. Describe a mood or name a title you loved. An agent searches the catalog, reads your taste, and calls a two-stage recommender. It can only re…

    Go

  3. kubernetes/kubernetes kubernetes/kubernetes Public

    Production-Grade Container Scheduling and Management

    Go 129k 45.1k

  4. vllm-project/vllm vllm-project/vllm Public

    A high-throughput and memory-efficient inference and serving engine for LLMs

    Python 96.7k 24.6k

  5. pytorch/pytorch pytorch/pytorch Public

    Tensors and Dynamic neural networks in Python with strong GPU acceleration

    Python 105k 30.6k

  6. opensearch-project/k-NN opensearch-project/k-NN Public

    🆕 Find the k-nearest neighbors (k-NN) for your vector data

    Java 230 254