Explore the full project graph in 3D
Every public GitHub repo, grouped by domain - Computer Vision, LLMs, Stats, MLOps, more.
Open
11 projects
AWS Redshift ML Internship - Summer 2026
Vector Search · Bedrock ML
Summer 2026 · AWS Redshift ML Team

AWS Redshift ML Infrastructure - Software Engineer Internship

Shipped native vector search and Bedrock ML integrations at Redshift scale. Won org-wide Amazon AUTA AI Hackathon (800+ interns) for a multi-model cost-aware router.

800+ interns · Winner 3.2x cost cut sub-100ms
AWS RedshiftAmazon BedrockVector SearchPythonSQL
View case study
Enterprise LLM evals schematic
Oct 2026 · LLM Evaluation

Enterprise LLM Evals: Three Evaluations Missing From The Public Corpus

Twenty enterprise scenarios in openai/evals format plus three novel evals the public corpus does not cover: tool-use trajectory, streaming-cancel JSON validity, and PII leakage through tool arguments. Cross-model: Sonnet 4.6, Opus 4.7, GPT-4.X.

20 scenarios 3 novel evals Cross-model
PythonAnthropic SDKopenai/evalsPrompt caching
View case study
Agent benchmark schematic
Sep 2026 · Benchmarks

Replit Agent Bench: Head-to-Head on 7 Realistic Dev Tasks

Reproducible benchmark comparing Replit Agent, Cursor paired with Claude Sonnet, and the raw Claude API across 7 real-world development tasks. One command to run, pinned model versions, open scoring rubric.

3 agents · 7 tasks make bench Quarterly reruns
PythonpytestAnthropic SDKDocker
View case study
Comet-shaped browser agent schematic
Sep 2026 · Browser Agents

Comet-Shaped Web Agent: Payload-Classified Permissions

Browser agent on Playwright and CDP, driven by Claude Sonnet 4.6 with tool use. Each proposed action classified by payload content before dispatch, so permission scope depends on what the agent is about to do, not just where.

Sonnet 4.6 Playwright + CDP Payload tiers
TypeScriptClaude Sonnet 4.6PlaywrightCDP
View case study
GSoC 2026 - Linux Foundation - Accord Project: APAP and MCP Server
GSoC 2026 · Linux Foundation · Accord Project

APAP/MCP Hardening - Google Summer of Code 2026

Selected as a Google Summer of Code 2026 contributor with The Linux Foundation (Accord Project). Service layer refactor, four-tier testing and multi-client docs for the APAP Model Context Protocol server.

175 hrs · 12 wks 53 tests · 98.55% cov 5 prior PRs
TypeScriptMCPDrizzle ORMPostgreSQLDocker
View case study
Driver Behavior Analytics
Dec 2024 · Bayesian Analytics

Driver Behavior Analytics System

Driver risk analytics for 300K+ drivers using survival analysis & Bayesian modeling, with real-time assessment APIs.

88% accuracy 10K+ daily req <200ms
KafkaPyMC3FastAPIPostgreSQL
View case study
LLM Serving Framework
May 2025 · LLM Infrastructure

Production LLM Serving Optimization

vLLM continuous batching, INT8/INT4 quantization and multi-GPU tensor parallelism for enterprise-scale inference.

12.3K req/s 42ms P50 70% memory cut
vLLMFastAPIK8sPrometheus
View case study
Search Relevance Ranking
Dec 2024 · Information Retrieval

Search Relevance & Ranking System

Production search engine with LambdaMART ranking across 500K+ queries, plus a comprehensive A/B testing framework.

nDCG@10 0.847 +18.3% CTR <200ms P95
XGBoostscikit-learnFlaskBM25
View case study
AgentForge Multi-Agent RAG
Nov 2024 · Multi-Agent RAG

AgentForge Multi-Agentic AI RAG

Multi-agent RAG with LangGraph orchestration for autonomous research, document generation and intelligent retrieval.

98.5% success −67% cost 180ms P95
LangChainLangGraphGPT-4Redis
View case study
Face Mask Detection
Nov 2023 · Computer Vision

Face Mask Detection

Real-time face mask detection using deep CNN with transfer learning for automated safety compliance.

99.6% accuracy <50ms inference 500+ concurrent
TensorFlowVGG16Flask
View case study