Builder · Bangalore, India
Portfolio · Email · LinkedIn · GitHub
I build systems that operate under uncertainty across finance, healthcare, security, and physical simulation. My work combines machine learning, backend engineering, and physics-based modeling to create systems that are measurable, explainable, and grounded in real-world constraints.
Operating principles: verify (look inside, not only at outputs), simulate (stress failure modes before reality costs), refuse (calibrated abstention when evidence is weak).
Domains I ship in: LLM interpretability / emotion probes, multi-host threat hunting & campaign correlation, quantitative trading research, clinical risk scoring, and drone CFD / scientific ML.
Anima — LLM emotion & uncertainty probes
Open-source instrumentation for Hugging Face causal LMs. Forward hooks + probe heads read valence, arousal, and uncertainty from hidden states per token, with a guard that can recommend abstaining. FastAPI / WebSocket streaming + dashboard. HF Spaces demo.
Keywords: LLM interpretability, emotion probing, Hugging Face, PyTorch, FastAPI.
Corvex — multi-host campaign correlator
Research correlator that fuses weak per-host detectors (lateral auth, micro-exfil, recon fanout) into ATT&CK-shaped attack timelines via HMAC-signed event envelopes. No LLM, no cloud API. Standing claim: holds up on sealed synthetic fleets; not validated on real enterprise telemetry or pure-benign baselines yet. Observe-only Windows / macOS sensors; live containment gated / dry-run.
Keywords: threat hunting, MITRE ATT&CK, lateral movement, SIEM-style correlation, cybersecurity, Python.
GeoQuant — quantitative trading platform
FastAPI + PyTorch stack for ML daily/intraday signals, news sentiment, walk-forward backtesting with costs in the loop, and Alpaca paper-trade routing.
Keywords: algorithmic trading, quantitative finance, backtesting, sentiment analysis, Alpaca.
Drift — clinical risk + health tracking
Flask health platform whose runtime path uses published ACC/AHA Pooled Cohort and FINDRISC models (not invented scores), plus a 17-biomarker timeline, wearables OAuth, and hard gates when data is incomplete.
Keywords: clinical risk, health tracker, biomarkers, explainable ML, Flask.
Self-driven CFD and simulators used to design aircraft I physically build and fly (plus a 10-day IISc vortex-ring internship, May 2025 — the repos are personal work that predates and continues independently of IISc):
| Repo | What it is |
|---|---|
| Drone-Vortex-Ring-Simulation | Reduced-order vortex rings — Kelvin Γ, Helmholtz self-induction, viscous decay |
| Propeller-simulator | BEMT propeller performance sweeps (GUI + CLI + CSV) |
| vortex-tracker | OpenCV ring diameter / propagation speed from high-speed imagery |
| NeuralVortex | ML surrogate (TFNO / Conv3D path) on own CFD volumetric fields |
Keywords: vortex ring, drone aerodynamics, CFD, fluid dynamics, Fourier Neural Operator, BEMT.
text2sql-rag (clean-room Spider text-to-SQL RAG) · VidhiSethu (Indian-legal RAG architecture docs) · homelab-rpi · cursor-llm-council
Summer Intern — Indian Institute of Science (IISc), Bangalore · May 2025 · 10 days
Vortex-ring internship; built vortex-tracker. CFD / simulation portfolio is self-driven, not an IISc research fellowship.
Engineering Intern — Vegam Solutions · Q3 2025 – Q1 2026
Air-filtration engineering (Q3); text-to-SQL RAG pipeline (Q1, NDA). Portfolio proof: text2sql-rag on the public Spider benchmark.
Founder — Athera · Ongoing
AI automation, lead workflows, and websites for small businesses.
Co-founder — ORQIS AI · Ongoing
Detect silent agent failures (e.g. tool-call loops), explain them, and open reviewable patches with a human in the loop.
Python · TypeScript · PyTorch · Hugging Face · FastAPI · Next.js · Flask · Postgres · Qdrant · Redis · Docker · Linux · Raspberry Pi
Competitive skating, football (CBSE / Goa Globe), badminton, swimming · Model UN · 50+ hrs community service (including NGO web presence).