Maulik Jain profile photo 1 on Tal

Maulik Jain

Senior Product Engineer at Fibr AI

Greater Bengaluru Area

8+ years of work experience

Fibr AI
Aug 20251 yr 1 moPresent

Bengaluru, Karnataka, India

• Architected the Ads Personalization agentic flow end-to-end, generating AI-driven change recommendations across page elements that lifted CVR by ∼10-11% over manual variants; built with prompt engineering, strict response-schema validation, iterative hyperparameter and prompt tuning driven by production evals, and checkpoint pause/resume for human-in-the-loop refinement. • Built Fibr Genesis, an AI landing-page generator, from POC to production, cutting draft time from ∼2 days to ∼5 minutes per page, with SSE token streaming, an in-editor chat agent for iterative refinement, and resilient reconnect handling for partial-event scenarios. • Built the CRO Audit flow end-to-end, a multi-agent pipeline that audits a prospect’s site for conversion bottlenecks, surfaces user segments via a Segment Discovery Agent, and turns the findings into runnable experiments through the campaign creation engine, cutting audit cycle time from ∼2-3 days of consultant work to ∼13 minutes. • Stood up the LLM gateway and observability stack on LiteLLM, Portkey, and Langfuse: centralised API-key handling at the gateway (no credential distribution to client services), provider-level model fallbacks, and unified LLM observability across the AI platform. • Built the Demo Sandbox for prospect onboarding, a reverse-proxy SDK-injection layer that previews Fibr on a prospect’s live site, with a parallel variant-generation workflow producing multiple personalised previews simultaneously. • Owned the supporting platform for these AI workloads, including Temporal Workflows orchestration, HPA + KEDA autoscaling, RBAC across services, and the AWS-to-GCP migration of dependent infrastructure.
Certa.ai
Sep 20231 yr 11 mosJul 2025

• Spearheaded Design Agent, a multi-agentic workflow management system that lets end-users modify JSON DSL workflows via natural language, integrating RAG and tool-calling with a feedback loop and error-tracking memory across calls; cut workflow setup from ∼2 months of admin time to a couple of days. • Engineered Risk Agent, the AI-powered assessment layer that surfaces and scores third-party vendor risks by analysing infosec documents (SOC2, ISO, HIPAA, PCI-DSS) against the SIG framework, cutting SIG response time from ∼1.5 weeks to ∼20-30 minutes per vendor. • Developed and deployed the Adjudication Agent, the decisioning layer that determines which surfaced risks are material under client-specific risk appetites, across compliance, security, and performance dimensions. • Provided technical leadership to the AI product team, driving ideation, architecture, and full-lifecycle delivery of AI-driven products from concept to production. • Architected the extraction of a core communication and collaboration feature from a monolith into a scalable, multi-tenant Python microservice on AWS Lambda + MongoDB.
PermuteHQ
Nov 20211 yr 10 mosAug 2023

• Led delivery of web applications across Fin-Tech, Agri-Tech, Prop-Tech, and Insure-Tech domains; shipped 10+ projects start-to-finish. • Built a vendor financing platform unlocking working-capital cash flows for SMBs in Python/Django/Svelte. • Designed an OCR + PDF metadata extraction pipeline that improves insurance-intermediary efficiency by precisely identifying customer, insured-entity, and premium data. • Built a customised job scheduler for batch processing across many data sources, with failover, failure tracking, and retries powering insights and recommendations.
RenewBuy
May 20192 yrs 6 mosOct 2021

Gurugram

• Overhauled the insurance-quote engine backbone, simplifying insurer onboarding and reducing integration effort by 20%. • Deployed centralised logging and APM across the distributed system, identifying and resolving critical bottlenecks for a ∼25% reduction in AWS spend. • Automated CI/CD for internal microservices using GitLab CI, Docker, and AWS spot instances; migrated legacy bash-script and BuildBot deployments onto the new architecture. • Built a system for point-in-time querying of internal hierarchies powering MIS, commissions, and incentives.
RenewBuy
Sep 20171 yr 8 mosApr 2019

• Built an OAuth-based authentication and authorisation system from scratch, acting as the central repository for encrypted PII and KYC documents across the platform. • Created a configuration-driven OLTP backend supporting both static and API-based pricing, standardising order placement through a generic data and API model with payment-gateway adaptors.

Maulik’s AI Stack

anthropic apis
anthropic apis
claude code
claude code
claude
claude
gemini
gemini
langfuse
langfuse
openai
openai

Education

NIIT University

Bachelor of Technology, Computational Science

NIIT University

20134 yrs2017

Join Maulik and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Maulik Jain on Tal

Headline

Senior Product Engineer at Fibr AI

Location

Greater Bengaluru Area

AI stack

anthropic apis

claude code

claude

gemini

langfuse

openai

Work experience

Senior Product Engineer at Fibr AI, Aug 2025 - Present · 1 yr 1 mo, Bengaluru, Karnataka, India, • Architected the Ads Personalization agentic flow end-to-end, generating AI-driven change recommendations across page elements that lifted CVR by ∼10-11% over manual variants; built with prompt engineering, strict response-schema validation, iterative hyperparameter and prompt tuning driven by production evals, and checkpoint pause/resume for human-in-the-loop refinement. • Built Fibr Genesis, an AI landing-page generator, from POC to production, cutting draft time from ∼2 days to ∼5 minutes per page, with SSE token streaming, an in-editor chat agent for iterative refinement, and resilient reconnect handling for partial-event scenarios. • Built the CRO Audit flow end-to-end, a multi-agent pipeline that audits a prospect’s site for conversion bottlenecks, surfaces user segments via a Segment Discovery Agent, and turns the findings into runnable experiments through the campaign creation engine, cutting audit cycle time from ∼2-3 days of consultant work to ∼13 minutes. • Stood up the LLM gateway and observability stack on LiteLLM, Portkey, and Langfuse: centralised API-key handling at the gateway (no credential distribution to client services), provider-level model fallbacks, and unified LLM observability across the AI platform. • Built the Demo Sandbox for prospect onboarding, a reverse-proxy SDK-injection layer that previews Fibr on a prospect’s live site, with a parallel variant-generation workflow producing multiple personalised previews simultaneously. • Owned the supporting platform for these AI workloads, including Temporal Workflows orchestration, HPA + KEDA autoscaling, RBAC across services, and the AWS-to-GCP migration of dependent infrastructure.

Softwares Development Engineer - III at Certa.ai, Sep 2023 - Jul 2025 · 1 yr 11 mos, • Spearheaded Design Agent, a multi-agentic workflow management system that lets end-users modify JSON DSL workflows via natural language, integrating RAG and tool-calling with a feedback loop and error-tracking memory across calls; cut workflow setup from ∼2 months of admin time to a couple of days. • Engineered Risk Agent, the AI-powered assessment layer that surfaces and scores third-party vendor risks by analysing infosec documents (SOC2, ISO, HIPAA, PCI-DSS) against the SIG framework, cutting SIG response time from ∼1.5 weeks to ∼20-30 minutes per vendor. • Developed and deployed the Adjudication Agent, the decisioning layer that determines which surfaced risks are material under client-specific risk appetites, across compliance, security, and performance dimensions. • Provided technical leadership to the AI product team, driving ideation, architecture, and full-lifecycle delivery of AI-driven products from concept to production. • Architected the extraction of a core communication and collaboration feature from a monolith into a scalable, multi-tenant Python microservice on AWS Lambda + MongoDB.

Technical Architect at PermuteHQ, Nov 2021 - Aug 2023 · 1 yr 10 mos, • Led delivery of web applications across Fin-Tech, Agri-Tech, Prop-Tech, and Insure-Tech domains; shipped 10+ projects start-to-finish. • Built a vendor financing platform unlocking working-capital cash flows for SMBs in Python/Django/Svelte. • Designed an OCR + PDF metadata extraction pipeline that improves insurance-intermediary efficiency by precisely identifying customer, insured-entity, and premium data. • Built a customised job scheduler for batch processing across many data sources, with failover, failure tracking, and retries powering insights and recommendations.

Senior Software Engineer at RenewBuy, May 2019 - Oct 2021 · 2 yrs 6 mos, Gurugram, • Overhauled the insurance-quote engine backbone, simplifying insurer onboarding and reducing integration effort by 20%. • Deployed centralised logging and APM across the distributed system, identifying and resolving critical bottlenecks for a ∼25% reduction in AWS spend. • Automated CI/CD for internal microservices using GitLab CI, Docker, and AWS spot instances; migrated legacy bash-script and BuildBot deployments onto the new architecture. • Built a system for point-in-time querying of internal hierarchies powering MIS, commissions, and incentives.

Software Engineer at RenewBuy, Sep 2017 - Apr 2019 · 1 yr 8 mos, • Built an OAuth-based authentication and authorisation system from scratch, acting as the central repository for encrypted PII and KYC documents across the platform. • Created a configuration-driven OLTP backend supporting both static and API-based pricing, standardising order placement through a generic data and API model with payment-gateway adaptors.

Education

Bachelor of Technology, Computational Science, NIIT University