Bijon Lahiri

Software Engineer

Hyderabad, Telangana, India15 yrs 9 mos experience

AI ML PractitionerAI Enabled

Key Highlights

Over 10 years of experience in AI/ML and Telecom Data Engineering.
Achieved 97% classification accuracy in 5G telemetry analysis.
Expert in building Databricks-based analytics platforms.

Stackforce AI infers this person is a Telecom Data Engineer specializing in AI/ML solutions for network optimization.

Contact

Skills

Core Skills

Data EngineeringAi For TelecomTelecom Engineering

Other Skills

Python (Programming Language)PySparkDatabricksDelta LakeLCELVector StoreApache SparkDatabricks ProductsApache Spark Streamingdelta live tableAmazon S3PlotlyKPI Dashboards5GNRORAN

About

I am an AI/ML and Telecom Data Engineer with over 10 years of experience bridging the gap between deep-layer telecom engineering (PHY/MAC) and modern cloud-scale AI. My expertise lies in building end-to-end diagnostic systems and Databricks-based analytics platforms for 4G/5G RAN networks. What I do: AI for Telecom: I develop ML pipelines (CNNs, LSTMs) and AI Agents for automated Root Cause Analysis (RCA), achieving up to 97% classification accuracy in 5G DU telemetry. Data Engineering: I architect Databricks Lakehouse environments using Medallion architecture (Bronze-Silver-Gold) and Spark Structured Streaming to handle 10GB/hour telemetry flows. Innovation: Recently, I’ve been building LLM-driven RCA agents using RAG (Retrieval-Augmented Generation) and LangChain to allow engineers to query network faults in natural language. Tech Stack: AI/ML: PyTorch, Scikit-learn, MLflow, Time-Series Modeling. Data: Databricks, Spark, Delta Lake, SQL, Python. Telecom: 4G/5G RAN KPIs (BLER, CQI, MCS, RI), L1/L2 Systems. I am passionate about making telecom networks more "self-healing" through proactive AI-driven automation.

Experience

15 yrs 9 mos

Total Experience

3 yrs 11 mos

Average Tenure

Current Experience

Radisys

2 roles

Staff Engineer

Jul 2024 – Oct 2025 · 1 yr 3 mos · Bengaluru, Karnataka, India · On-site

Telecom KPI RCA AI Agent
Designed and implemented an AI-driven Root Cause Analysis (RCA) agent for telecom KPIs (BLER, CQI, MCS, RI, throughput) to accelerate fault diagnosis in 4G/5G RAN environments.
Built an interactive Dash-based UI enabling engineers to query RCA insights in natural language and receive structured, human-readable diagnostic summaries.
Integrated the agent with a Delta Lakehouse (Bronze–Silver–Gold) architecture on Databricks, sourcing RCA context from curated KPI tables with CDC-driven historical tracking.
Implemented context-aware RCA reasoning by correlating time-aligned KPIs (e.g., BLER spikes vs CQI degradation) across cells, UEs, and sites.
Designed the foundation for retrieval-augmented generation (RAG) by planning vectorization of KPI metadata, historical anomalies, and domain heuristics to improve RCA accuracy.
KPI Dashboard Analytics Platform — Telecom Network Data Lakehouse
Modelled and designed an enterprise-grade OLAP Lakehouse on AWS S3 using Databricks, PySpark, Delta Lake, and Spark Declarative Pipelines to support large-scale KPI analytics across multi-vendor 5G field deployments.
Architected end-to-end Bronze → Silver → Gold pipelines with append-Vlow streaming ingestion to consolidate heterogeneous telecom logs from production Vield sites and lab test environments.
Implemented auto-CDC with SCD Type-2 to preserve historical state for audit readiness and to enforce standardized KPI computation and schema governance across all analytics consumers.
Built real-time RCA (Root-Cause Analysis) dashboard layers using correlation-driven event classification, reducing RCA turnaround time by 40% and enabling proactive anomaly detection for Vield deployments.
Improved downstream analytics reliability by ~40%, reduced ingestion latency by ~30%, and fully automated incremental refresh processing supporting both batch and streaming workloads at scale.

Python (Programming Language)PySparkData EngineeringAI for Telecom

Lead Engineer

Feb 2020 – Jun 2024 · 4 yrs 4 mos · Bengaluru, Karnataka, India · On-site

Automation & Streaming Data Pipeline for Live Anomaly Detection of O-RU
Designed spark streaming pipelines for real time health monitoring of RU
Processed telemetry (EVM, EPRE, frequency offset) at more than 10GB/hour scale
Implemented checkpointing, window aggregation, and SLA driven alerting
Reduced manual validation effort by 60% through proactive signal monitoring
Minimal Impact on Communication KPIs
Built anomaly detection for L1/L2 mismatches (TTI stretch/miss, slot mismatch).
Integrated Spark jobs with L2 CLI workflows for automated cell recovery.
Developed observability dashboards tracking recovery time and SLA metrics.
Optimized ingestion using parallel ETL and Delta checkpoints, improving latency by 35%.
Developed Python automation and KPI monitoring frameworks.
Built data instrumentation pipelines improving throughput by 9.8 Mbps.
Processed multi-GB test logs into structured analytics layers.

Python (Programming Language)PySparkData EngineeringAI for Telecom

Mediatek

3 roles

Staff Engineer

Promoted

Jul 2019 – Feb 2020 · 7 mos · Greater Bengaluru Area

Involved in the design and implementation of DUAL RADIO DUAL STANDBY (DR-DSDS) for LTE
Extensive design discussions of LTE scheduler for proper and efficient scheduling of secondary sim
Designed and developed LTE scheduler and PDCCH mimo control state machine for accommodation of dual sim support
Incorporated interface modifications between L1 driver and INNER receiver to accommodate for seamless integration
Analysis and debugging of live issues on LTE
Lead a team of 10 members for analysis and debugging of live issues related to WCDMA
Responsible for mentoring junior team members for ramp up on WCDMA technology and RAKE receiver implementation
Paper on “Dual Radio Dual Sim Receiver for LTE Applications”

Telecom Engineering

Senior Engineer

Promoted

Jul 2017 – Jun 2019 · 1 yr 11 mos · Greater Bengaluru Area

Contributed in the design and implementation of EPDCCH feature
Implementation of changes for EPDCCH done for control channel MIMO module
UT and IT verification for control channel MIMO demodulation module related to EPDCCH feature
Suggested and implemented calculation of performance metrics of control channel MIMO demodulation module for easier debugging of issues related to PDCCH CRC fail
Module ownership for control channel MIMO demodulation and noise estimation based on cell specific reference signal
Created scripts using Python for parsing and plotting of key parameters related to RAKE receiver for easier issue analysis of live issues in WCDMA
Lead a team of 4 members for analysis and debugging of live issues related to WCDMA
Responsible for mentoring junior team members for ramp up on WCDMA technology and RAKE receiver implementation

Telecom Engineering

Engineer

Jul 2016 – Jun 2017 · 11 mos · Greater Bengaluru Area

Responsible for implementation of channel estimation related HW blocks of RAKE receiver on vector dsp for optimised area implementation
Involved in the implementation of low power receiver for WCDMA
Analysis and debugging of real time issues related to WCDMA specifically low power receiver feature
Paper on “Low Power Receiver Implementation for WCDMA”

Telecom Engineering

Vishveya systems pvt. ltd.

Member Technical Staff

Jul 2015 – Jun 2016 · 11 mos · Greater Bengaluru Area

Consulting for Saankhya Labs
Understanding WRAN 802.22 standards
Analyze the feasibility of implementation on target DSP processor
Development and integration of individual fixed point C and Assembly modules of uplink, downlink, transmitter and receiver PHY layer
Feature addition of transmit power control in downlink path

Telecom Engineering

Indian institute of technology, bombay

2 roles

Research Scholar

Nov 2011 – Jun 2015 · 3 yrs 7 mos · Mumbai Metropolitan Region

Presently, I'm doing research in the field of Piezoelectric Energy Harvesting at Dept. of Electrical Engg. IIT Bombay under the supervision of Prof. V.R. Rao

Master of Technology (M. Tech.)

Jul 2009 – Oct 2011 · 2 yrs 3 mos · Mumbai Metropolitan Region

I did my Masters specializing in Microelectronics from IIT Bombay. Working in a project on Deployment of Wireless Sensor Networks for Explosive Detection.