About | Manish Botta

I’m a software engineer based in San Francisco. My work lives where backend engineering, applied AI, and real customer problems meet.

I’ve built retrieval and agent systems inside Microsoft’s sales organization, worked with enterprise teams at Yellow.ai, and owned product and engineering for TradeLiv as its first and sole engineer.

I enjoy the parts that make a system useful beyond a demo: permission checks, thoughtful data models, reliable queues, and a clear path from a customer’s problem to something they can use. I’m comfortable moving between architecture, implementation, and the conversations that shape both.

Experience

Apr 2026
to Present

TradeLiv

Founding Software Engineer (product and engineering)

First hire and sole engineer; product discovery through full-stack delivery.

Delivered a B2B SaaS platform from zero to production in four months, spanning 40+ REST endpoints and designer, administrator and client workflows, by owning product discovery, architecture and full-stack implementation as the first hire and sole engineer.
Reduced new e-commerce integration time from two days to under 15 minutes by replacing per-site scrapers with a reusable AI-assisted product-extraction pipeline combining browser rendering and structured product parsing.
Enabled end-to-end procurement across product curation, client approval, checkout and brand-specific purchase orders by implementing the React/TypeScript frontend, Express APIs and PostgreSQL/Prisma data model.
Enabled clients to review designer shortlists and receive live updates without repeated page refreshes by implementing token-based client portals and scoped server-sent event streams.

TypeScriptReactViteNode.jsExpressPostgreSQLPrismaStripePuppeteerAI-assisted extractionSSE

Aug 2025
to Apr 2026

Microsoft

Software Engineer (Contract)

Owned retrieval, orchestration and reliability components within a team-built field-sales agent. Overall agent: 75,000 daily active users. Owned component/request category: 1.5 million requests over six months.

Delivered retrieval and orchestration components processing 1.5 million requests over six months within a field-sales agent serving 75,000 daily active users by implementing and extending workflows across CRM, document and SQL data sources on Azure.
Improved retrieval-context relevance by approximately 30–40% by replacing fixed-size chunks with heading-based parent/child chunks that preserved context across long contract sections.
Supported new sales-agent use cases while retaining context across tool calls by implementing LangGraph nodes, edges, shared state and routing, and extending the existing orchestration graph.
Bounded agent planning and tool-execution failure paths with timeouts, capped retries, exponential backoff and fallbacks by implementing explicit controls and dead-letter handling for stalled or unsuccessful runs.
Delivered an RLHF-trained model into production by implementing reward-model training and the reinforcement-learning stage using human feedback.

PythonAzureLangGraphLangChainLangSmithPostgreSQLSQLPineconeBGE/Qwen embeddingsRAGrerankingRLHFCRM integrations

Sep 2021
to Jul 2023

Yellow.ai

Software Engineer

Backend/services and enterprise delivery for the first nine months, followed by product engineering. Worked with 25+ enterprise customers on a platform encompassing approximately 50 integrations.

Delivered enterprise chatbot solutions for 25+ customers by translating customer requirements into platform features, implementing integrations and coordinating launches with product and customer-success teams.
Improved intent-classification accuracy by 35% relative to the baseline and reduced unrecognized inputs or fallback rates by 60% in owned components by improving intent/entity handling and adapting open-source models with LoRA.
Improved CSAT by 35% relative to the baseline by strengthening query routing across FAQs, document knowledge and automated workflows, with nearest-match suggestions and live-agent handoff when required.
Reduced API response time from 820 ms to 490 ms by optimizing MySQL, PostgreSQL and MongoDB queries and caching hot reads in Redis.

Node.jsPythonKafkaMySQLPostgreSQLMongoDBRedisBERTspaCyLoRAintent classificationentity recognitionconversational AICRM integrations

Mar 2021
to Aug 2021

Cognizant

Program Trainee Analyst, Full Stack Developer

Internal banking proof of concept in a team delivery environment; hands-on service development, testing, deployment and monitoring.

Reduced response time from 1.2 seconds to 700 ms in an internal banking proof of concept by helping decompose a Java monolith into four Spring Boot microservices.
Achieved 85% code coverage in the proof of concept by writing unit tests and carrying out regression, white-box and black-box testing.
Enabled independent deployment and operational monitoring of four services by deploying them on AWS and configuring CloudWatch monitoring and automated alerts.

JavaSpring BootSpring MVCH2MySQLAWS EC2S3Elastic BeanstalkCloudWatch

Tools I work with

Backend & systems

Python
TypeScript
Java
FastAPI
Flask
Node.js
Spring Boot
PostgreSQL
Redis
Kafka
WebSockets
SSE

AI & retrieval

LangGraph
LangChain
LangSmith
RAG
Parent/child chunking
Embeddings
Reranking
BERT
spaCy
LoRA
Reward modeling
RLHF
Input/output guardrails

Data & delivery

MySQL
MongoDB
SQLite
Azure
AWS
OCI
Docker
GitHub Actions
Nginx
OAuth 2.0
RBAC

Education

M.S. Computer Science

California State University, San Bernardino

Aug 2023 – May 2025

B.Tech Computer Science

Amrita Vishwa Vidyapeetham, Coimbatore, India

Jul 2017 – May 2021

How I work

Start with the real workflow

Talk to the people using the system. Understand their constraints before committing to an architecture.

Make the failure paths work

Retries, permissions, observability, and recovery belong in the design from the beginning.

Own the last mile

Deployment, integrations, documentation, and production support are part of building the product.