Nishanth B - PortfolioI build systems where AI outputs have consequences.
$> Verification Systems_
Validation > Generation
// SYS.MODULE_INIT
$> I'm an AI Infrastructure Engineer building verification-first systems where AI outputs must survive real consequences.
Smart contracts. Financial workflows. Operational decisions. When mistakes are expensive, generation isn't enough. Validation matters.
Founded NexOps (BCH-1 Hackcelerator winner, Top 5 finalist Cardano Hackathon Asia). Building Clawback Labs.
Former SDE Intern at TVS ASL, open-source contributor at BrainGlobe, and builder of reliable systems used by real users. Strong in Python, distributed systems, DSA and AI.
SDE experience across enterprise and open-source systems.
// SDE Intern
// Open Source Contributor
MODULE_MATRIX // V.2.0.4 — languages, frameworks, and verification tooling.
[ Languages ]
[ Frameworks ]
[ Databases ]
[ ML / AI ]
[ Tools ]
[ Engineering ]
[ Focus ]
Projects that scaled — verification-first systems in production.
NexMCP — Security-First AI Framework for Bitcoin Cash
AI-assisted CashScript engineering with multi-phase validation (Skeleton → Logic → Validation). AST + semantic checks across 50+ deployed contracts — used by 3 external teams and 20+ developers.
Breaking API changes, fixed in your CI
Open-source GitHub Action that watches providers your SaaS depends on, finds breaking-change call sites, and opens a reviewable fix PR. Source stays in CI — no code sent to an LLM.
AI-Native Founder Operating System
AI-native founder OS coordinating agents for research, planning, memory, and decisions — RAG, pgvector, model routing, and deterministic guardrails instead of blind autonomy.
Governed Multi-Agent Creation and Automation Platform
Ops console that turns Knowledge Graphs into scoped agents — Temporal orchestration, MCP + OAuth tools, Redis limits, and evidence-first PASS / FAIL / NO_EVIDENCE validation.
AI-Powered Quantum Computing & Analytics Platform
Random Forest backend recommender over 10K+ quantum jobs — FastAPI + Qiskit + PostgreSQL dashboards for success rates, execution performance, and queue efficiency.
AI-assisted invoice auditing
Invoice auditing that reconciles contracts against vendor bills and surfaces billing discrepancies before payment — explainable verification, not opaque predictions.
SYS.ACHIEVEMENT_STREAM :: ACTIVE — recognition from competitions and awards.
Built NexOps and won the overall BCH-1 Hackcelerator, receiving the $10,000 grand prize.
Prize winner ($800) for building blockchain infrastructure tooling.
Winner at ETHnile for blockchain / infrastructure work.
Placed Top 5 among 3,500+ registrations and 200+ finalist teams.
Top 10 among 700+ teams at AQVH'25.
Principles that shape how I build high-consequence systems.
Most AI products optimize for generation. The real challenge is verification.
Users don't care whether a workflow is agentic. They care whether it works.
Reliability is a product feature. Not an engineering metric.
Benchmarks reveal reality. Demos reveal potential.
The most useful AI systems are not the smartest. They're the most dependable.
Field notes from shipping verification systems under real constraints.
While building NexOps I learned that most generation failures weren't caused by model intelligence. They were caused by ambiguous specifications. Once the contract intent became structured, model quality mattered much less than validation quality.
A good validator catches errors that clever prompts can't. A structured spec validator is the real leverage point.
With HelmOS I realized the hard part isn't planning or execution. It's knowing when they're wrong and recovering gracefully.
Demos hide edge cases. Benchmarks expose them. If you can't measure it, you don't understand it.
The real bottleneck isn't generation. It's knowing whether the output is correct. That's where the effort belongs.
Your validation system should be interpretable to a human. Black box verification means users can't debug failures. Explainability compounds trust.
Get in contact — interested in developer tools, verification systems, and AI reliability.