hey, i'm

Vaibhav Hariram

building paprika — runtime governance for ai agents. trace capture, policy enforcement, deterministic replay. deployed to 25+ engineering teams.

prev. swe intern @ railinc (summer '24 + '25). react + java/spring boot on the RIGIS platform. etl pipeline processing 100K+ rail stations.

studying cs @ berkeley

fine-tuned 7B–13B LLMs on 1.2M samples, distributed training across 4×A100s for berkeley's gorilla project (BAIR). cited by nvidia, anthropic, openai teams.

researcher @ uc berkeley (eecs), satellite imagery + ML to detect crypto mining development in el salvador (Potts Lab)

personal projects

routeX — routing engine that does in <1ms what dijkstra does in 45. contraction hierarchies over 280K bay area road segments, postgis geofencing, xgboost eta predictions.

stack: c++ • postgis • xgboost • osm

swandb — feature store that actually respects time. pit-correct joins, <8ms online serving, parquet storage that doesn't lie to your model.

stack: duckdb • redis • parquet

chronicle — llm inference server — 32% throughput over hf baseline. micro-batching + kv-cache reuse. <200ms p95 at 100 concurrency.

stack: python • cuda • mistral-7b

snare — (see repo)

open source

vLLM / PyTorch — merged performance optimization + distributed benchmarking PRs upstream

geofilter — numba jit point-in-polygon engine. 10M points in 0.58s. wrote it because existing spatial filters were too slow and i was impatient.

reqbench — load-testing cli sustaining 4.8K req/sec at 500 concurrency with p95/error ci gating

lore

pretty good at geography trivia

played trevor rainbolt in ranked geoguessr. got absolutely cooked tho

played carnatic violin for 10 years

huge chelsea fan + chicago sports

grew up in cary, north carolina. escaped.

tl;dr

i love geospatial ai/ml + geotech: geospatial data, machine learning, and real systems people actually use

currently

berkeley → scale and build wherever and whatever i can

still thinking about that caleb williams throw…

some other things

coursework: data structures, algorithms, discrete math, structure of computer programs, a.i., machine learning, networking + internet protocols, database systems, computer vision, data science, probably more ai/ml

languages: python, c++, typescript, java, sql

ml + systems: FSDP, AMP, CUDA, KV-cache, distributed fine-tuning, eval pipelines, quantization

frameworks: react, fastapi, spring boot, node.js, duckdb, langgraph, pytorch, xgboost

infra: aws, docker, redis, postgresql, postgis, gdal, arcgis pro, linux, ci/cd

interests: cities + public transportation, collecting old maps, tamil culture, soccer jerseys, geoguessr, chicago sports, international music

less frequently: south indian movies, home decor, sketching, online chess, sparkling water

other things

used to watch anime a little bit

trying to learn to dj and cook (or both at the same time)

will run a marathon this year

trying to attend more hackathons

roast my spotify playlists