I'm a senior software engineer with 10+ years of experience.
Today I focus on Agentic AI systems β multi-agent pipelines that do real-world research, forensic analysis, and automation, not just chat. I particularly love the engineering challenge of making LLM-powered systems reliable, not just impressive.
When I'm not shipping code, I teach high school mathematics and enjoy breaking down complex technical concepts into things anyone can understand.
π― My superpower: I can navigate between a low-level circuit diagram and a high-level LLM orchestration graph without losing context. Rare combo.
π΅οΈ ForensicValue AIMulti-agent forensic accounting for Indian stocks Detects accounting irregularities, governance risks & value traps in NSE/BSE companies. Five parallel agents (Forensic, Management, RPT, Market Intelligence, Narrative) orchestrated via LangGraph, synthesized into a "Detective Story" report.
|
Daily autonomous research pipeline for special-situation investing Monitors NSE/BSE filings for insider trades & corporate events, then runs a 4-level deep-research pipeline (DeepSeek V3 + Qwen). Persists insights in PostgreSQL with
|
π€ Autonomous SRE AgentAI that reads your logs so you don't have to β at 3 AM Ingests production logs, clusters them with HDBSCAN, flags anomalies with Isolation Forest, investigates root causes via cross-encoder reranked RAG, and drafts fixes for human approval. Human-in-the-loop learning improves recommendations over time.
|
π Medicine FinderAI-powered drug information assistant A conversational tool for looking up medicines, active ingredients, and substitutes β built with a clean, fast React frontend and an intelligent backend that surfaces the right information quickly and safely.
|
PRs to projects I depend on. Sorted by repo stars, descending. (merged) = in upstream, (open) = in review.
- vercel/next.js
β140,087 β
fix 4 broken links in app router guides and example README;fix broken documentation links in examples and templates;honor pageExtensions in no-html-link-for-pages(open) - denoland/deno
β107,151 β
update Chocolatey install link to community.chocolatey.org(merged β );fix dead MDN link to Body interface(open) (open) - vitejs/vite
β81,510 β
fix broken rolldown plugin options link in api-plugin.md;fix broken and malformed documentation links(open) - OpenHands/OpenHands
β77,472 β
fix typo in OpenHands docs URL in add_agent.md(merged β ) - withastro/astro
β60,240 β
remove dead llms.txt reference(merged β ) - microsoft/autogen
β59,029 β
fix 2 broken external doc links (FAQ + Aspire dashboard)(open) - remix-run/react-router
β56,441 β
fix 13 broken api.reactrouter.com/v7/functions/ reference links(open) - BerriAI/litellm
β50,665 β
apply cache_control_injection_points on /v1/messages route;prevent top-level 'type' from leaking into OpenAI function parameters;apply default_team_params.models to newly created teams;fix broken guardrails docs link in ibm_guardrails README;fix broken links in README and enterprise docs(open) - pixijs/pixijs
β47,430 β
remove dead https://pixijs.io/customize link in v7 migration guide(open) - mattermost/mattermost-mobile
β2,678 β
disable syntax highlighting and line numbers for code blocks without language;date separator missing day of month on certain Android devices(open) - lidatong/dataclasses-json
β1,484 β
respect default_factory for abstract generic fields in from_dict(open) - The-OpenROAD-Project/ORAssistant
β41 β
Fix 5 bugs: PDF crash, unused env var, missing state field, hardcoded model, typo;Fix 4 runtime bugs: crash, streaming None, discarded replace, memory leak(open)
- Building more reliable agentic systems (the harness matters as much as the model)
- Forensic accounting automation for Indian public markets
- Low-level systems β revisiting the ECE roots
- Value investing through the lens of AI-assisted research