Anurag Nagare's Work | ContraWork by Anurag Nagare
Anurag Nagare

Anurag Nagare

I’m an AI & Machine Learning engineer with expertise in deve

    1x
    Hired
    5
    Followers
Followed by Michael L, Baby ganesh G, and GALLERY L

Mumbai, India

WorkPostsServicesAbout

Mumbai, India

Neuro Lend My credit model scored 0.92 AUC. Then I changed one field on one applicant, female to male, and she went from approved to rejected. A lot of lending AI works like this: it says no in milliseconds, and nobody, not the customer and often not the bank, can say why. So I built NeuroLend. You fill in a loan application and watch the signal travel through every neuron until the output turns green or red. Then it shows why it decided, and the smallest change that would flip the answer. Rohan is rejected at 54% risk. If he cuts his other EMIs to ₹5,500 a month, or asks for ₹7.5 lakh instead of ₹9 lakh, he's approved. Here's what's happening underneath. It's a small 8 → 10 → 6 → 1 network trained on 40,000 synthetic Indian loan applications, where I deliberately gave gender zero effect on the outcome. The "why" uses Integrated Gradients against an average applicant instead of raw activations, because glowing neurons look great but explain nothing to a human. Counterfactuals come from a binary search on each input the applicant can actually change, and a fairness check re-scores every applicant with only gender flipped. That check caught the bias that accuracy missed. Retraining with gender cut out of the network dropped the effect to exactly zero, and AUC stayed at 0.92. It all runs in plain JavaScript in the browser, about 0.005 ms per decision, with no server. Code is in the comments. How does your team test lending models for bias today? Medium - https://medium.com/@anuragnagare77/neuro-lend-99cddd19faad GitHub - https://github.com/AnuragNagare/Neuro-Lend.git
0
29
CryptoHorizon an interactive app that makes the quantum threat to encryption real instead of theoretical: a live Shor's-algorithm demo, a classical-vs-post-quantum crypto benchmark, a personal "harvest now, decrypt later" risk calculator, and a sourced Q-Day timeline, all in one dashboard. Defining feature is honesty by design: every number on screen is computed live at request time, never hardcoded RSA-2048 and ECC P-256 keys get generated, benchmarked, and compared against NIST's ML-KEM and ML-DSA in real time, and the quantum demo runs an actual Qiskit circuit against small, pre-validated semiprimes rather than a lookup table dressed up as a result. Only one thing is allowed to leak into the UI's language: a rule baked in from commit one that nothing may imply real-world RSA is broken or claim quantum advantage. Under the hood, Shor's algorithm runs for real superposition over a counting register, controlled modular exponentiation, inverse QFT, continued-fraction extraction, then a classical gcd step to pull out the factors on a FastAPI backend with no shortcuts. The risk calculator checks your required confidentiality window against five independently sourced Q-Day estimates from NIST, Google, and the Global Risk Institute, so "safe until 2035" is a real, cited comparison, not a guess. medium.com/@anuragnagare77/cryptohorizon-i-built-a-quantum-computer-that-factors-91-and-thats-the-whole-point-2c156a00d0da
0
113
One Gate: Voice-controlled Gmail, Calendar, and Drive that can't fire the wrong thing Every "AI agent" demo shows a voice command turning into a sent email like magic, but almost nobody shows the part that actually matters: what stops it from sending the wrong thing. I built proof: an agent that runs your Gmail, Calendar, Contacts, and Drive by voice reply to a thread, schedule a meeting, archive an email, find a file hands-free end to end, but never fires anything irreversible without you saying so. The honest hard part isn't getting an LLM to sound smart, it's this: a transcript goes to LLM against a strict JSON schema and comes back as an ordered plan, every step tagged reversible or not and exactly one thing in the whole system is allowed to check that flag. Finding a thread, drafting a reply, creating a calendar event: those just run. The result covers real ground without ever feeling like it's guessing: forward or reply to email with the original quoted underneath, archive/label/trash, resolve a name to a real address through your Contacts first and your mail history as fallback, schedule an event that creates quietly and only emails the invite after a second confirmation, answer "what did John say" or "what's on my calendar Thursday" grounded in content actually fetched from your account not hallucinated. No backend, no server anywhere in the loop
0
87
What every "AI gesture control" demo quietly leaves out Everyone films a hand waving at a screen and calls it AI. Almost nobody shows what's underneath that there's usually no gesture model at all. A "grab" is one distance crossing a line. Here's proof. I built a jigsaw puzzle you solve with your bare hands no mouse, no controller, no gesture classifier, no training.
0
148