System online
AI / Software Engineer
I build intelligent systems,
developer infrastructure,
and reliable backend software.
Next /experience
Jun 2026 — Present
Software Development Engineer Intern
Nov 2025 — Mar 2026
Researcher & Developer
Reconstruction pipeline: a single uncalibrated image is turned into a depth field with DepthAnything, camera pose is recovered with DUSt3R, the scene is reconstructed as a sparse point set, and that set is resolved into 3D Gaussian splats.
Input
Monocular
Depth
DepthAnything
Camera pose
DUSt3R
Reconstruction
Gaussian splat
3DGS
Sep 2025 — Nov 2025
Business & Product Associate
Nov 2024 — Feb 2025
AI Engineer Intern
Autonomous software engineering system
A multi-stage pipeline that carries software from requirements through architecture, code generation, static checks, QA and real test execution, with human review gates and evidence-driven retry loops between the stages. Typed Pydantic contracts, isolated run workspaces, structured logging, provider fallback and token-budget enforcement keep long runs predictable; failure-aware retry, degrade and abort strategies decide what happens when a stage does not pass. A FastAPI service with SQLite persistence and WebSocket updates drives the run, executes and lints the generated code, and verifies it against an automated test suite.
Autonomous SDLC
✓Passed/Stage 06/06
test_executionok
runcomplete
Execution pipeline / LangGraph / Python
Revenue recovery and auditability system
A revenue recovery agent, built for the Razorpay Buildathon, that can prove what it did, what it spent and why it stopped. A deterministic policy engine — not the model — owns every spend decision, governed by stopping rules, a consent registry, retry caps, a cost ceiling, expected-value checks and quiet hours. Every outcome is committed to a SHA-256 hash-chained, append-only ledger with tamper detection, and refusals are recorded exactly like actions are. Gemini diagnoses unrecognised errors, reads inbound replies and drafts customer copy, with deterministic fallbacks. Measured against a 20% holdout across 2,000 contacts: +33.9 pp incremental lift, 95% CI +33.0 to +34.9. Razorpay payment links and settlement webhooks run in test mode; customer outcomes and sends are simulated.
Recovery control
▲Tamper detected
Representation of the RecoverOS recovery decision system. A failed payment is ingested and classified, then a deterministic policy engine decides whether to act or refuse. Acting proceeds through a recovery channel; refusing terminates in a recorded reason code. Both outcomes are committed to a SHA-256 hash-chained, append-only audit ledger, and a tamper check demonstration detects a modified entry.
PolicyCost ceiling/Exceeded→Refuse
Ledger / SHA-256 / append-only
▲Tamper detected at 003
001
7f3a1c
payment.failed002
b2e084
policy.act003
5d9c47
settle.recorded004
a10f6b
policy.act005
c48e2d
policy.refuse134Automated tests
+33.9 ppIncremental lift95% CI +33.0 to +34.9
Razorpay test mode / outcomes simulated
Reinforcement learning for bug discovery
A microservice codebase modelled as a connected 18-node, 36-edge Watts–Strogatz graph, where a tabular Q-learning agent learns to walk between modules and test them for bugs. The reward function pays +50 for a critical bug and +10 for a minor one while charging −1 per step, so the policy has to trade coverage against effort; state is discretised to (node, time bucket, tested bucket, already-tested flag), which is what stops the agent oscillating between two high-value modules. The agent is benchmarked against a random-walker baseline on identical episode seeds. Around it sits the MLOps: MLflow experiment tracking, a DVC pipeline driven by a single params file, drift and reward-regression monitoring, versioned policy snapshots with rollback, and a Streamlit dashboard. The published evaluation reports 152.59 average reward against the baseline’s 94.42 — +57.9% reward — at a 97.94% bug-discovery rate over 100 episodes.
RL environment
✓Episode complete
Reinforcement-learning experiment. A microservice codebase is modelled as a connected 18-node, 36-edge Watts-Strogatz graph of modules such as Auth, Database and API Gateway. A tabular Q-learning agent starts at Auth and moves to a neighbouring module each step, testing it for bugs. Discovering a critical bug pays +50, a minor bug +10, and every step costs -1, so the agent is pushed to find bugs in as few moves as possible. The recorded episode shown runs 22 steps and finds all 6 seeded bugs for a total reward of 118.
Reward
+50 critical / +10 minor / -1 step
Policy
Tabular Q-learning / α 0.1 / γ 0.95 / ε 1.0→0.05
Offline experiment / recorded episode
18 nodes / 36 edges / Watts–Strogatz
I'm a final-year B.E. student in Artificial Intelligence and Machine Learning at BMS College of Engineering, Bangalore. Most of what I build sits where models meet infrastructure — multi-agent systems, retrieval pipelines, gateways, and the guardrails that make them safe to run in production.
Right now I work on cloud infrastructure at Skylo, where I built a centralized MCP gateway and the org-wide safety rules that constrain what AI agents are allowed to do. Before that I researched calibration-free 3D reconstruction at Samsung R&D, shipped a production RAG system at HiDevs, and worked the product side of a seed round at NeoSapien.
I also lead technical programs for GRADIENT, the AI club at BMSCE — 100+ members, five hands-on workshops and two hackathons on LLMs, computer vision and agentic workflows.
Email is the fastest way to reach me. If it's about a system, send the problem — I'd rather read that than a pitch.