⌘ observatory

Recruiter

Arch

Command center·Active systems 3·Stream 10K+ events / day·Pipelines LIVE

Yash Jadhav
Full Stack + GenAI Systems Engineer

Building production-ready GenAI pipelines & real-time AI systems

Read positioning— specialization & how I work

React, WebSockets, automation pipelines, vector databases, and real-time intelligence platforms. Engineering control-room mindset: shipping interfaces, streaming backends, and automation that move latency, accuracy, and throughput — not slide decks.

Or jump to the live dashboard for systems and metrics.

Collaboration console

Portrait

Live signals

GitHub activity unavailable (rate limit or offline)Systems shipped: 3Stream 10K+ events / dayPipelines LIVE

control.metrics

Live systems dashboard

Quantified outcomes with architecture rationale on every metric — the control-room view recruiters scan first.

Outcome lines on each card; stack / mechanism detail on hover (desktop) or expandable note (mobile / touch).

0K+

Market events / day

Operators see live market throughput without hammering APIs on every tick.

Hover or focus — how it’s built

0M+

Signals processed

Analysts query history and scores without reconstructing the full raw tape each time.

Hover or focus — how it’s built

0%

Latency reduced

Faster time-to-signal on the live operator surface vs. traditional pull-only dashboards.

Hover or focus — how it’s built

10 hrs/wk

Automation reclaimed

Roughly a day a week back for humans vs. manual download, packaging, and babysitting jobs.

Hover or focus — how it’s built

0%

Research accuracy lift

Tutoring and internal research answers stay better grounded and repeatable.

Hover or focus — how it’s built

github.public

systems.explorer

Systems architecture explorer

Shipped systems with pipelines, explicit decisions, and proof links — not anonymous bullet lists.

Problem → pipeline → tech → decisions. Expand each system to walk the control-room narrative; architecture mode tightens diagram density.

Open full case study

Problem

Operators need sub-second market structure awareness without drowning in noise or stale dashboards.

Constraints

Sub-second signal delivery to operators — REST polling cannot keep up with tick granularity.
Multi-source ingestion and fusion without alert fatigue on mobile channels.
Low operational overhead while iterating quickly on scoring and UI.
Durable signal history for replay, audits, and downstream analytics.

Architecture — sketch, pipeline, tech

Architecture sketch

Pipeline

Input

Exchange & feed adapters
WebSocket ingress
Telegram / operator channels

Process

Fan-out + backpressure
Confidence scoring
Mapbox density layers
Signal persistence

Output

Live map + dashboards
Ranked alerts
Historical replay

Tech & surface

WebSocket fan-out, market feed adapters (e.g. Alpaca-class APIs), Telegram bot alerts, Mapbox density layers, Supabase for 1M+ stored signals, confidence scoring, paper-trading simulation surface.

Engineering decisions

WebSockets over polling
Primary driver: persistent streaming for sub-second awareness. Co-benefit: avoids REST polling storms and cuts redundant load on the same tick stream.
Supabase for signal store
Fast iteration on relational + realtime primitives without standing up a bespoke data plane first.
Confidence-ranked UI
Surfaces signal quality so operators triage by trust, not raw volume.

Tradeoffs

Managed data plane

Supabase over self-hosted Postgres-only or MongoDB-first

Pros

+Realtime subscriptions and auth-aligned storage out of the box
+Faster ship path for hackathon / product iteration

Cons

−Vendor dependency vs. full control of every infra knob
−Deep custom tuning lives behind provider limits

What broke & how it was fixed

Burst fan-out pressure
Symptom Early paths struggled when concurrent clients and bots subscribed to the same high-rate stream.
Response Introduced bounded queues and explicit backpressure so ingest could shed load without corrupting state.

Evolution

v1
REST-style polling prototype to validate dashboard interest and tick shape.
v2
WebSocket streaming ingress with fan-out to dashboards.
v3
Confidence scoring layer, Mapbox density, Telegram path — operators triage by trust, not volume.

Impact

10K+ market events/day processed; 1M+ signals stored; ~40% latency reduction vs REST-heavy paths; geospatial trading behaviour on Mapbox; ranked buy/sell signals and Telegram alerting.

Decision timeline

2023Polling prototype validated operator demand for live structure
2024 H1WebSocket ingress + fan-out replaced REST polling storms
2024 H2Scoring, Mapbox density, Telegram — triage by confidence not volume

Proof & media

link

Live NeuroTicker demoVisit

Public deployment — end-to-end real-time surface.

Live preview

signal.recognition

Recognition & wins

Filter by signal type, then expand any card for story and proof links.

6 shown

Spotlight

Global Top 12

International Top 12 — Duke Kunshan University

Cross-border competition, research-grade bar

Story & media— expand

Placed among the top teams internationally — validation on a global stage, not just local circuits.

Winner

HackaNova Hackathon '26 — Winner

Built and pitched under extreme time pressure

Story & media— expand

First place at HackaNova '26 — end-to-end product thinking, live demo discipline, and clear storytelling when judges ask “why this architecture?”

image

Set src on this media item in lib/site.ts

Add `src` in lib/site.ts after placing a file under public/portfolio/ (e.g. `/portfolio/hackanova.jpg`).

video

Add href (YouTube or Loom) in lib/site.ts.

Set `href` to YouTube or Loom for this win.

Winner

Avishkar '24 — Winner (RITE)

National innovation & engineering showcase

Story & media— expand

Winner at Avishkar (RITE) — recognition for technical depth, execution, and presentation in a national forum.

Finalist

Aptos Hackathon '25 — Finalist

Web3 + systems at hackathon speed

Story & media— expand

Finalist at Aptos '25 — shipping on-chain-adjacent flows and developer UX fast enough to survive a weekend sprint.

Finalist

Level Supermind Hackathon '25 — Finalist

AI / cognition–adjacent product lane

Story & media— expand

Finalist at Level Supermind '25 — competing where product sense and model-backed UX both matter.

Winner

GCP & GenAI Labs '23 — Winner

Google Cloud GenAI track + peer mentoring

Story & media— expand

Winner in GCP & GenAI Labs '23 — applied GenAI patterns with cloud-native discipline; also mentored peers through the same labs (leadership + teaching signal).

link

Verify credential (GiveMyCertificate)Visit

Official GCP GenAI Labs completion verification.

link

GitHub profile & projectsVisit

Code and repos tied to shipped systems.

Verify credential

Winner

HackaNova Hackathon '26 — Winner

Built and pitched under extreme time pressure

Story & media— expand

First place at HackaNova '26 — end-to-end product thinking, live demo discipline, and clear storytelling when judges ask “why this architecture?”

image

Set src on this media item in lib/site.ts

Add `src` in lib/site.ts after placing a file under public/portfolio/ (e.g. `/portfolio/hackanova.jpg`).

video

Add href (YouTube or Loom) in lib/site.ts.

Set `href` to YouTube or Loom for this win.

Winner

Avishkar '24 — Winner (RITE)

National innovation & engineering showcase

Story & media— expand

Winner at Avishkar (RITE) — recognition for technical depth, execution, and presentation in a national forum.

Finalist

Aptos Hackathon '25 — Finalist

Web3 + systems at hackathon speed

Story & media— expand

Finalist at Aptos '25 — shipping on-chain-adjacent flows and developer UX fast enough to survive a weekend sprint.

Finalist

Level Supermind Hackathon '25 — Finalist

AI / cognition–adjacent product lane

Story & media— expand

Finalist at Level Supermind '25 — competing where product sense and model-backed UX both matter.

Winner

GCP & GenAI Labs '23 — Winner

Google Cloud GenAI track + peer mentoring

Story & media— expand

Winner in GCP & GenAI Labs '23 — applied GenAI patterns with cloud-native discipline; also mentored peers through the same labs (leadership + teaching signal).

link

Verify credential (GiveMyCertificate)Visit

Official GCP GenAI Labs completion verification.

link

GitHub profile & projectsVisit

Code and repos tied to shipped systems.

Verify credential

experience.work

Work experience

R&D ownership across model lifecycle, APIs, and automation — metrics over slides.

Full Stack Developer (React + GenAI): scalable, user-focused production apps and real-world systems — not prototypes.

Dec 2024 – Jan 2026

R&D Intern

STSARC Pvt. Ltd.

ReactFlaskFastAPIRAGSLM routingVectorDBREST APIsMLOps

Built Algorithm — AI DSA trainer (React + Flask) with production SLM deployment, fine-tuning, and session-aware dynamic routing APIs
~30% faster student query resolution through retrieval + SLM feedback loops
~50% improvement in research accuracy on internal / tutoring research flows
Automation and tooling that reclaimed ~10 hrs/week and tightened feedback loops
End-to-end ownership from model lifecycle to APIs — measurable gains, not slide-only AI

R&D internship as one continuous thread: model lifecycle, APIs, and automation that show up as measurable outcomes — not slide-only AI.

Algorithm (React + Flask + SLMs) is the flagship here: RAG-grounded answers, session-aware routing, and deployable models in the same production path I owned end to end.

link

GitHub — projects & reposVisit

Code behind Algorithm, streaming work, and automation tooling.

background.education

Education

Formal CS foundation alongside shipping production systems.

2023 – 2027 (pursuing)

B.E. Information Technology

DY Patil College of Engineering, Pune

CGPA 8.03 — core CS: DSA, OOP, DBMS, OS, CN

Formal IT with strong CS cores; transcripts sit next to production engineering, not in place of it.

The same semesters that cover DSA, databases, and networks are the ones where hackathons and shipped systems earned external signal.

link

LinkedIn — education & winsVisit

Programs, competitions, and public timeline alongside coursework.

background.involvement

Involvement

Community and long-horizon commitments beyond pure engineering — consistency and ownership in the round.

Aug 2023 – Present

Community Development Volunteer

Bhaktivedanta Book Trust

Long-horizon community work alongside engineering — consistency and ownership outside pure tech.

Multi-year volunteer commitment — the same ownership muscle as shipping software, applied to community operations and people-facing work.

It is not a separate persona; it is how I practice reliability when the timeline is years, not sprints.