Shivam Jaiswal profile photo 1 on Tal

Shivam Jaiswal

Full-Stack Engineer . AI Systems . Multi-Agent Pipelines . Distributed Systems . Enterprise Dev Tooling . Founder @Tracktions

Greater Delhi Area

4+ years of work experience

Tracktions
Mar 20242 yrs 6 mosPresent

Pune District, Maharashtra, India

Founded and built Tracktions — a Trading Operating System for Indian retail traders — as a solo founder-developer, owning every layer from product decisions to production infrastructure. Built on Go, PostgreSQL (AWS RDS), Angular, and AWS (EC2, S3), the platform delivers 8 integrated modules covering strategy rule adherence, behavioral analytics, automated risk management, circuit breakers, and a phase-based profitability progression system. Made deliberate cost-vs-performance architecture decisions suited for early-stage SaaS: SSE over WebSockets for async notifications, PostgreSQL heavy indexing over Redis for read-heavy operations, and in-memory queuing over Kafka for a non-distributed system — each with a clear upgrade path as scale demands. Key outcomes: sub-100ms dashboard load times, 10,000+ concurrent user capacity, 80% reduction in manual trade logging effort, and real-time P&L tracking with live open trade monitoring. Currently live at www.tracktions.com with founding members onboarded and actively growing.
BNY Mellon
Jul 20224 yrs 2 mosPresent

Pune District, Maharashtra, India

Building AI systems and developer tooling at enterprise scale. → Architected an internal AI developer tooling marketplace: a locally-hosted, secure distribution system for configurable AI skills and MCP server configurations. Enables developers to self-serve AI tooling inside BNY's security perimeter — one-click install, automated config injection, contribution framework for team-built skills. Reduced repetitive DevOps and QA tasks by 60%. → Designed and shipped a multi-agent AI orchestration pipeline: two LLM-powered agents (Acceptance Criteria Parser + Canonical Message Generator) automating UAT validation across 30+ datasets and 50+ microservices. Zero manual intervention per cycle, eliminating ~40 hours of testing effort. → Built 9 production AI skills (DB querying, log analysis, application scaling, code review, Jira integration, test execution) and 8 automated workflows — packaged in a version-controlled sharing framework with automated setup scripts. → Led frontend for a critical data platform: new architecture, coding standards, strict review culture — 70% bug reduction while shipping major features over 6 months. → Authored 1,100+ lines of structured LLM prompts with field-extraction logic and a custom DSL-based validation schema supporting 7 distinct validation types.
HackDev
Jun 20182 mosJul 2018

Noida, Uttar Pradesh, India

Shivam’s AI Stack

anthropic apis
anthropic apis
antigravity
antigravity
claude code
claude code
claude
claude
codex
codex
cursor
cursor

Education

Delhi Technological University (Formerly DCE)

Bachelor of Technology - BTech, Computer Engineering

Delhi Technological University (Formerly DCE)

20184 yrs2022
Delhi College of Engineering

Bachelor of Technology - BTech, Computer Engineering

Delhi College of Engineering

20184 yrs2022

Join Shivam and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Shivam Jaiswal on Tal

Headline

Full-Stack Engineer . AI Systems . Multi-Agent Pipelines . Distributed Systems . Enterprise Dev Tooling . Founder @Tracktions

About

As a Full Stack Developer 2 at BNY Mellon, I focus on designing and implementing AI-driven systems and developer tooling within an enterprise-scale environment. I led the architecture of an internal AI developer tooling marketplace to enhance productivity and simplify configuration processes, resulting in a significant reduction of repetitive DevOps and QA tasks. Additionally, I contributed to a multi-agent AI orchestration pipeline that automated UAT validation across numerous datasets and microservices, eliminating the need for manual intervention. As the founder of Tracktions, I independently built a trading journal SaaS tailored for Indian retail traders. Leveraging technologies like Go, PostgreSQL, and AWS, I developed features such as strategy adherence, behavioral analytics, and automated risk management. Holding a BTech in Computer Engineering from Delhi Technological University, I am passionate about solving complex problems by creating innovative systems and architectures. Currently, I am exploring opportunities in senior AI engineering or founding engineer roles on a global scale.

Location

Greater Delhi Area

AI stack

anthropic apis

antigravity

claude code

claude

codex

cursor

Work experience

Founder at Tracktions, Mar 2024 - Present · 2 yrs 6 mos, Pune District, Maharashtra, India, Founded and built Tracktions — a Trading Operating System for Indian retail traders — as a solo founder-developer, owning every layer from product decisions to production infrastructure. Built on Go, PostgreSQL (AWS RDS), Angular, and AWS (EC2, S3), the platform delivers 8 integrated modules covering strategy rule adherence, behavioral analytics, automated risk management, circuit breakers, and a phase-based profitability progression system. Made deliberate cost-vs-performance architecture decisions suited for early-stage SaaS: SSE over WebSockets for async notifications, PostgreSQL heavy indexing over Redis for read-heavy operations, and in-memory queuing over Kafka for a non-distributed system — each with a clear upgrade path as scale demands. Key outcomes: sub-100ms dashboard load times, 10,000+ concurrent user capacity, 80% reduction in manual trade logging effort, and real-time P&L tracking with live open trade monitoring. Currently live at www.tracktions.com with founding members onboarded and actively growing.

Full Stack Developer 2 at BNY Mellon, Jul 2022 - Present · 4 yrs 2 mos, Pune District, Maharashtra, India, Building AI systems and developer tooling at enterprise scale. → Architected an internal AI developer tooling marketplace: a locally-hosted, secure distribution system for configurable AI skills and MCP server configurations. Enables developers to self-serve AI tooling inside BNY's security perimeter — one-click install, automated config injection, contribution framework for team-built skills. Reduced repetitive DevOps and QA tasks by 60%. → Designed and shipped a multi-agent AI orchestration pipeline: two LLM-powered agents (Acceptance Criteria Parser + Canonical Message Generator) automating UAT validation across 30+ datasets and 50+ microservices. Zero manual intervention per cycle, eliminating ~40 hours of testing effort. → Built 9 production AI skills (DB querying, log analysis, application scaling, code review, Jira integration, test execution) and 8 automated workflows — packaged in a version-controlled sharing framework with automated setup scripts. → Led frontend for a critical data platform: new architecture, coding standards, strict review culture — 70% bug reduction while shipping major features over 6 months. → Authored 1,100+ lines of structured LLM prompts with field-extraction logic and a custom DSL-based validation schema supporting 7 distinct validation types.

Internship Trainee at HackDev, Jun 2018 - Jul 2018 · 2 mos, Noida, Uttar Pradesh, India

Education

Bachelor of Technology - BTech, Computer Engineering, Delhi Technological University (Formerly DCE)

Bachelor of Technology - BTech, Computer Engineering, Delhi College of Engineering