Setu — Pre-emptive discipline layer for AI coding agents
2026 – present

Most AI coding agents fail the same way: they hallucinate project conventions, ignore existing abstractions, and produce code that passes tests but breaks the codebase's internal contract. Setu sits in front of the agent and prevents this before generation.

It uses a JIT Context Engine to surface the most relevant files, patterns, and constraints at inference time, and a DAG Swarm to parallelise multi-file reasoning across the project graph.

TypeScriptJIT Context EngineDAG SwarmLLM Agents
github.com/pkgprateek/setu-opencode
RAG Arena 2026 — Competitive evaluation framework for RAG pipelines
2026

RAG benchmarks are broken. Most evaluate one configuration in isolation, obscuring what actually drives quality.

RAG Arena runs configurations head-to-head on the same queries using an LLM-as-judge panel for faithfulness, relevance, and groundedness. Pluggable retrievers, pluggable corpora, live leaderboard.

PythonFastAPIChromaDBLLM-as-judge
github.com/pkgprateek/rag-arena-2026
SmartData Lab — Distributed training infrastructure for large-scale ML
2024–25

Led the redesign of the distributed training stack for multi-node GPU clusters. The bottleneck was inter-node communication during backward passes.

The fix combined FSDP sharding, NCCL tuning, and FP16 → FP8 mixed precision. Wall-time dropped 3.2× on the same hardware.

3.2× training speedup Multi-node A100 clusters
PyTorchFSDPNCCLSlurmFP8CUDA
YVO Service — High-throughput checkout API and ML-backed personalisation
2023–24

The checkout flow was timing out at ~3k RPS. Rewrote the core API in Hono, moved product reads to a Redis-backed edge cache, and decoupled the recommendation service into an async fire-and-forget.

Peak throughput reached 12k RPS. Cart abandonment dropped 65%.

12k RPS peak 65% abandonment reduction
HonoTypeScriptRedisPostgreSQL