René's URL Explorer Experiment


Title: Ankit Jha | Staff AI Engineer

Open Graph Title: Ankit Jha | Staff AI Engineer

X Title: Ankit Jha | Staff AI Engineer

Description: Ankit Jha is a Staff AI Engineer with 7 years of experience building AI systems end to end: agents, retrieval and evaluation, down to LLM inference, request routing and the infrastructure underneath.

Open Graph Description: Ankit Jha is a Staff AI Engineer with 7 years of experience building AI systems end to end: agents, retrieval and evaluation, down to LLM inference, request routing and the infrastructure underneath.

X Description: Ankit Jha is a Staff AI Engineer with 7 years of experience building AI systems end to end: agents, retrieval and evaluation, down to LLM inference, request routing and the infrastructure underneath.

Mail addresses
ankitjha.email@gmail.com
ankitjha.email@gmail.com
ankitjha.email@gmail.com

Opengraph URL: https://pyextreme.github.io/

direct link

Domain: pyextreme.github.io


Hey, it has json ld scripts:
{"@context":"https://schema.org","@type":"Person","name":"Ankit Jha","jobTitle":"Staff AI Engineer","url":"https://pyextreme.github.io","sameAs":["https://github.com/PyExtreme","https://www.linkedin.com/in/ankitjha-42"],"email":"mailto:ankitjha.email@gmail.com"}

theme-color#111816
og:typewebsite
og:site_nameAnkit Jha
og:imagehttps://pyextreme.github.io/og.png
og:image:width1200
og:image:height630
og:image:altAnkit Jha, Staff AI Engineer. End-to-end AI systems: agents, retrieval, evaluation, inference and infrastructure.
twitter:cardsummary_large_image
twitter:imagehttps://pyextreme.github.io/og.png

Links:

Skip to contenthttps://pyextreme.github.io#main
Ankit Jhahttps://pyextreme.github.io/
Projectshttps://pyextreme.github.io/projects/
Open sourcehttps://pyextreme.github.io/open-source/
Abouthttps://pyextreme.github.io/about/
GitHub ↗https://github.com/PyExtreme
See the workhttps://pyextreme.github.io/projects/
Open-source contributionshttps://pyextreme.github.io/open-source/
Why a recreated AIBrix pod inherited the old pod's blacklist blockhttps://pyextreme.github.io/open-source/aibrix-3523/
All projects →https://pyextreme.github.io/projects/
01MarketLane AIAI-native market intelligence for NSE equities: how the daily scan went from one cached LLM call to a verified multi-stage pipeline, and why the middle generation ships.Product2026PythonFastAPI↗https://pyextreme.github.io/projects/marketlane-ai/
02MarketLane ResearchEvidence-grounded retrieval over SEC filings, and the measurement system that decided what shipped: 13 pre-registered experiments, two adopted changes, eleven rejected with mechanisms, about $40 of model spend.Infrastructure2026Python 3.12FastAPI↗https://pyextreme.github.io/projects/marketlane-research-rag/
03AI Incident CommanderA pre-registered evaluation of agent architectures for automated incident diagnosis — single agent, multi-agent commander, Bayesian causal engine, and HolmesGPT (CNCF) — across 150 held-out incidents on two domains, eight frozen experiments, and ~1,280 runs for $51.Research2026PythonClaude Haiku↗https://pyextreme.github.io/projects/ai-incident-commander/
04Agent Context EngineA Python library that decides what a long-running AI agent puts in its model context on every call: three memory layers, hybrid retrieval, four compression methods and a rule-based adaptive planner, benchmarked against Claude on 2,204 graded answers.Infrastructure2026Python 3.11SQLite↗https://pyextreme.github.io/projects/agent-context-engine/
All 24 contributions across 9 repositories →https://pyextreme.github.io/open-source/
BerriAI/litellmhttps://github.com/BerriAI/litellm
Keeping carried end-user spend when LiteLLM retries a budget resethttps://pyextreme.github.io/open-source/litellm-50898/
Write-uphttps://pyextreme.github.io/open-source/litellm-50898/
PR ↗https://github.com/BerriAI/litellm/pull/51143
Issue ↗https://github.com/BerriAI/litellm/issues/50898
ai-dynamo/nixlhttps://github.com/ai-dynamo/nixl
An integer overflow that bypassed NIXL's registration boundshttps://pyextreme.github.io/open-source/nixl-2520/
Write-uphttps://pyextreme.github.io/open-source/nixl-2520/
PR ↗https://github.com/ai-dynamo/nixl/pull/2548
Issue ↗https://github.com/ai-dynamo/nixl/issues/2520
BerriAI/litellmhttps://github.com/BerriAI/litellm
Stop caching empty replies in LiteLLM's response cachehttps://pyextreme.github.io/open-source/litellm-49931/
Write-uphttps://pyextreme.github.io/open-source/litellm-49931/
PR ↗https://github.com/BerriAI/litellm/pull/50057
Issue ↗https://github.com/BerriAI/litellm/issues/49931
AI Incident Commanderhttps://pyextreme.github.io/projects/ai-incident-commander/
Agent Context Enginehttps://pyextreme.github.io/projects/agent-context-engine/
MarketLane Researchhttps://pyextreme.github.io/projects/marketlane-research-rag/
MarketLane AIhttps://pyextreme.github.io/projects/marketlane-ai/
vLLM #57650https://pyextreme.github.io/open-source/vllm-57650/
llm-d-inference-sim #801https://pyextreme.github.io/open-source/llm-d-inference-sim-801/
llm-d-router #2488https://pyextreme.github.io/open-source/llm-d-router-2488/
vLLM Router #299https://pyextreme.github.io/open-source/vllm-router-299/
LinkedIn ↗https://www.linkedin.com/in/ankitjha-42
GitHub ↗https://github.com/PyExtreme
GitHub ↗https://github.com/PyExtreme
LinkedIn ↗https://www.linkedin.com/in/ankitjha-42

Viewport: width=device-width, initial-scale=1


URLs of crawlers that visited me.