-:-:- · SURAT, IN
backend & systems engineersurat, india
Backend & Systems Engineer
ml, dl & llm harness implementation - inference, agents, mcp, on-device
Founding Engineer @ Saras - building the systems & AI behind advisor intelligence over live-market data.
Backend engineer who finds zen in the terminal. New tech drops and I go feral to reverse-engineer it. A 1000x extrovert with people I know, a 10000x introvert with everyone else. Big-time overthinker, terminal-over-IDE kinda guy.
BACKENDSYSTEMSAIINDIE OSS
now playing
nothing rn
offline · zzz
-:--
current stack
python
pytorch
vllm
go
postgres
cloudflare
typescript
recent writing
slate: Turning a Phone into a Stream Deck for macOS - One Command, a Pluggable Backend, and the Trade-offs Behind ItHow I built slate - a free, open-source phone-as-Stream-Deck for Mac that runs over your LAN with no cloud. One semantic Command, a pluggable Transport, a Zod schema shared across three languages, and the trade-offs (including why not WSS) behind it.Jul 202620 minNo Public Ports: Zero-Trust Team SSH to a Private Azure Cluster with a Reverse Bastion and Cloudflare WARPHow I gave a whole team SSH to a private Azure dev cluster with zero internet-facing ports - one outbound-dialed cloudflared bastion, Cloudflare WARP, and Google SSO - plus the split-tunnel routing trap I hit on the way.Jul 202617 minBear the Tokens: FP16 and Suffix Decoding to 9,000+ Tokens a Second on a Tesla T4How I pushed Qwen2.5-0.5B to 9,052 tok/s on a free Colab Tesla T4 in the H2Loop inference contest - from AWQ INT4 dead ends to FP16 plus model-free suffix speculative decoding.Jun 202621 minLogpipe: Building a Distributed Logging Stack from Scratch with Vector and ClickHouseHow I built Logpipe - a small but real Vector + ClickHouse logging pipeline in Go. Why batching at an aggregator is the whole point, what 'Too many parts' actually means, and how disk buffers stop you from losing logs on a restart.Jun 202625 min