curiosity.grit.execution
I want machines that remember the way people do — not as a pile of similar text, but as facts that begin, hold, and end.
Being
Surface answers bore me. If something catches, the next move is automatic: mechanism, then the system that produced it, then where it breaks, then what happens at scale, then what would replace it. I don't want 'that's interesting, tell me more.' I want the structure that generated the answer. Settled subjects can wait. Things that are still moving get my attention.
I don't trust a title. I also don't reject a title for sport. If the reasoning holds, I'll take the conventional line. If it doesn't, I will say so. Knowing-what is cheap. Understanding-why is the job. 'Industry standard' is a prompt, not a conclusion. First principles, not a personality brand.
I am not trying to collect prestige. Prestige is useful as an instrument. What I want is capability, optionality, leverage, rooms where the work is real, and ownership of direction. Achievement says 'I want to succeed.' I am asking how good I could become, and what I could build if I did. Safe-and-adequate is a baseline. The finish line moves. That is both the engine and the tax.
Curiosity without execution is a hobby. I would rather ship the next honest version than decorate a toy. I lift. I eat my way through Bangalore. Rhythm is not a footnote. I am restless: another layer, another frontier, another question. The honest risks are spreading too thin, over-optimizing decisions, getting impatient with slow compounding, and falling for frontier hype. High ambition only works if the skepticism stays loud.
Open edges
Temporal facts. Predicate-aware invalidation. Graph-native retrieval. The opposite of stuffing a context window and hoping.
I want systems that real people touch, under latency and trust constraints. Deeper ownership. Larger blast radius. Less slideshow, more product.
Ablations. Limits. Toy graphs labeled as toy. A statistical engine that stays deterministic even when a language model is invited to explain it.
Language first. Then agents that last. Then, if the memory is real enough, the world. I am not in a hurry to skip the middle.
Work exp
On the workbench
Version control for Excel — Git-shaped, spreadsheet-native, aimed at FP&A rather than developers. V0 is two people on one workbook: scenarios, review, cell-level merge. What exists now is M1: a round-trip .xlsx bridge so values, formulas, and basic formatting survive parse → write → parse. I do not claim a shipped collaboration product.
Aug 2026A toy query-conditioned DistMult NBFNet retriever, re-derived from G-reasoner. Typed entity–document graph in PyTorch Geometric. I say toy on purpose: ~20 documents, 0.45M params, CPU. On a curated multi-hop set the GNN recovered ranking (multihop MRR 0.14 → 0.82) against the same frozen encoder, with a no-edges ablation so the structure has to earn its keep.
Jun 2026A temporal knowledge graph memory for agents. Raw text becomes episodes, entities, and time-bounded facts in Neo4j. Exclusive predicates close; additive ones coexist. Retrieval is Cypher, not cosine. Inspired by Graphiti. V1 is library-first, no embeddings, no REST. The missing piece I actually want to spend years on.
Apr 2026A Python statistical library on PyPI. The tests are deterministic. Assumptions are checked. A language model may interpret the result, but it is not allowed to become the result. Independent t-test is the current MVP. The philosophy will outlive the test menu.
Query
Not me in the room. A chatbot speaking as me — opinions included. It should not invent a biography. It should not jailbreak.